Doctor-Shotgun commited on
Commit
61944bd
1 Parent(s): e89b36d

Initial model commit

Browse files
README.md ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: mistralai/Mistral-7B-v0.1
4
+ tags:
5
+ - generated_from_trainer
6
+ model-index:
7
+ - name: supercot-lora
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
15
+ # supercot-lora
16
+
17
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.9790
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 0.0002
39
+ - train_batch_size: 2
40
+ - eval_batch_size: 2
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 4
43
+ - total_train_batch_size: 8
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: cosine
46
+ - lr_scheduler_warmup_steps: 10
47
+ - num_epochs: 3
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss |
52
+ |:-------------:|:-----:|:----:|:---------------:|
53
+ | 0.7661 | 0.06 | 20 | 1.5173 |
54
+ | 0.7681 | 0.12 | 40 | 1.2323 |
55
+ | 0.6647 | 0.18 | 60 | 1.1306 |
56
+ | 0.6742 | 0.24 | 80 | 1.0847 |
57
+ | 0.6995 | 0.3 | 100 | 1.0573 |
58
+ | 0.6883 | 0.36 | 120 | 1.0412 |
59
+ | 0.6437 | 0.42 | 140 | 1.0375 |
60
+ | 0.6331 | 0.48 | 160 | 1.0186 |
61
+ | 0.6686 | 0.54 | 180 | 1.0153 |
62
+ | 0.6767 | 0.6 | 200 | 1.0042 |
63
+ | 0.7037 | 0.66 | 220 | 1.0023 |
64
+ | 0.6994 | 0.72 | 240 | 1.0014 |
65
+ | 0.7012 | 0.78 | 260 | 0.9996 |
66
+ | 0.6599 | 0.84 | 280 | 0.9926 |
67
+ | 0.6401 | 0.9 | 300 | 0.9913 |
68
+ | 0.6665 | 0.96 | 320 | 0.9910 |
69
+ | 0.5771 | 1.02 | 340 | 0.9907 |
70
+ | 0.6286 | 1.08 | 360 | 0.9830 |
71
+ | 0.6064 | 1.14 | 380 | 0.9865 |
72
+ | 0.5976 | 1.19 | 400 | 0.9802 |
73
+ | 0.5512 | 1.25 | 420 | 0.9817 |
74
+ | 0.6333 | 1.31 | 440 | 0.9810 |
75
+ | 0.5883 | 1.37 | 460 | 0.9817 |
76
+ | 0.5822 | 1.43 | 480 | 0.9783 |
77
+ | 0.5878 | 1.49 | 500 | 0.9757 |
78
+ | 0.5951 | 1.55 | 520 | 0.9753 |
79
+ | 0.6466 | 1.61 | 540 | 0.9719 |
80
+ | 0.6246 | 1.67 | 560 | 0.9681 |
81
+ | 0.627 | 1.73 | 580 | 0.9705 |
82
+ | 0.6214 | 1.79 | 600 | 0.9691 |
83
+ | 0.6558 | 1.85 | 620 | 0.9709 |
84
+ | 0.5736 | 1.91 | 640 | 0.9674 |
85
+ | 0.6188 | 1.97 | 660 | 0.9674 |
86
+ | 0.5293 | 2.03 | 680 | 0.9742 |
87
+ | 0.5463 | 2.09 | 700 | 0.9766 |
88
+ | 0.5184 | 2.15 | 720 | 0.9776 |
89
+ | 0.5349 | 2.21 | 740 | 0.9783 |
90
+ | 0.5536 | 2.27 | 760 | 0.9794 |
91
+ | 0.5016 | 2.33 | 780 | 0.9822 |
92
+ | 0.5075 | 2.39 | 800 | 0.9795 |
93
+ | 0.5529 | 2.45 | 820 | 0.9808 |
94
+ | 0.5168 | 2.51 | 840 | 0.9784 |
95
+ | 0.5416 | 2.57 | 860 | 0.9793 |
96
+ | 0.4845 | 2.63 | 880 | 0.9804 |
97
+ | 0.5487 | 2.69 | 900 | 0.9801 |
98
+ | 0.5313 | 2.75 | 920 | 0.9797 |
99
+ | 0.5449 | 2.81 | 940 | 0.9790 |
100
+ | 0.5303 | 2.87 | 960 | 0.9795 |
101
+ | 0.5599 | 2.93 | 980 | 0.9795 |
102
+ | 0.544 | 2.99 | 1000 | 0.9790 |
103
+
104
+
105
+ ### Framework versions
106
+
107
+ - Transformers 4.34.0.dev0
108
+ - Pytorch 2.0.1+cu118
109
+ - Datasets 2.14.5
110
+ - Tokenizers 0.14.0
adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 16,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 32,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj",
20
+ "q_proj",
21
+ "down_proj",
22
+ "o_proj",
23
+ "up_proj",
24
+ "gate_proj",
25
+ "k_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08b1b73d7017de9865ed46d6cd6c3a9a6127bd7eacb5a2d5b98d8c9252a1c6a0
3
+ size 335705741
added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<s>": 1,
4
+ "<unk>": 0
5
+ }
checkpoint-1002/README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: True
10
+ - load_in_4bit: False
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.6.0.dev0
checkpoint-1002/adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 16,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 32,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj",
20
+ "q_proj",
21
+ "down_proj",
22
+ "o_proj",
23
+ "up_proj",
24
+ "gate_proj",
25
+ "k_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
checkpoint-1002/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08b1b73d7017de9865ed46d6cd6c3a9a6127bd7eacb5a2d5b98d8c9252a1c6a0
3
+ size 335705741
checkpoint-1002/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1dc1ed40f6d09fcc28a2392cf8d4e2d24ef052999e08549c8bee62d25c628347
3
+ size 168624735
checkpoint-1002/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b29fa98b9f012bf405284a68bc9dc06a47b65fe7d90a86aba3fd58f90be1117
3
+ size 14575
checkpoint-1002/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b5ab1dfeb32f4a52cfa67b0aaf0fc080fca4001f56f082b5192b86d67057e0d
3
+ size 627
checkpoint-1002/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1002/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7780122e038bb83236a6b213b013867a89c4e742cce8c1208c1a41fc54e6c57
3
+ size 4475
checkpoint-334/README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: True
10
+ - load_in_4bit: False
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.6.0.dev0
checkpoint-334/adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 16,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 32,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj",
20
+ "q_proj",
21
+ "down_proj",
22
+ "o_proj",
23
+ "up_proj",
24
+ "gate_proj",
25
+ "k_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
checkpoint-334/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5361471ef99d12a2bd6efcc7a68a21d280f6054762518fed67fc416be2f01be5
3
+ size 335705741
checkpoint-334/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91e3312b7119b30229c6eb665c4eef218eab7199dccf42ffc62ef45dd5c3ec70
3
+ size 168624735
checkpoint-334/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28e2a0fb7ee32f0a066b898de199a4667218aa32e9df38aef2669bbfd6da825a
3
+ size 14575
checkpoint-334/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13d1c67cb4f0a266db63ac99579361487f41f1b07b1caa83a062b71665d0ef19
3
+ size 627
checkpoint-334/trainer_state.json ADDED
@@ -0,0 +1,2151 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9977595220313666,
5
+ "eval_steps": 20,
6
+ "global_step": 334,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 2e-05,
14
+ "loss": 1.5197,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4e-05,
20
+ "loss": 1.5435,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 6e-05,
26
+ "loss": 1.3967,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 8e-05,
32
+ "loss": 1.368,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.01,
37
+ "learning_rate": 0.0001,
38
+ "loss": 1.2766,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 0.00012,
44
+ "loss": 1.1676,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.02,
49
+ "learning_rate": 0.00014,
50
+ "loss": 1.0507,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.02,
55
+ "learning_rate": 0.00016,
56
+ "loss": 1.0016,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.03,
61
+ "learning_rate": 0.00018,
62
+ "loss": 0.9427,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.03,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.9294,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.03,
73
+ "learning_rate": 0.00019999949852874652,
74
+ "loss": 0.9023,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.04,
79
+ "learning_rate": 0.00019999799412001546,
80
+ "loss": 0.8257,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.04,
85
+ "learning_rate": 0.0001999954867888952,
86
+ "loss": 0.9064,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.04,
91
+ "learning_rate": 0.00019999197656053288,
92
+ "loss": 0.8408,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.04,
97
+ "learning_rate": 0.00019998746347013398,
98
+ "loss": 0.9058,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.05,
103
+ "learning_rate": 0.0001999819475629623,
104
+ "loss": 0.8419,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.05,
109
+ "learning_rate": 0.00019997542889433915,
110
+ "loss": 0.8535,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.05,
115
+ "learning_rate": 0.00019996790752964305,
116
+ "loss": 0.8121,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.06,
121
+ "learning_rate": 0.00019995938354430898,
122
+ "loss": 0.7725,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.06,
127
+ "learning_rate": 0.00019994985702382758,
128
+ "loss": 0.7661,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.06,
133
+ "eval_loss": 1.517298698425293,
134
+ "eval_runtime": 15.7157,
135
+ "eval_samples_per_second": 37.097,
136
+ "eval_steps_per_second": 18.58,
137
+ "step": 20
138
+ },
139
+ {
140
+ "epoch": 0.06,
141
+ "learning_rate": 0.0001999393280637444,
142
+ "loss": 0.7878,
143
+ "step": 21
144
+ },
145
+ {
146
+ "epoch": 0.07,
147
+ "learning_rate": 0.00019992779676965885,
148
+ "loss": 0.7435,
149
+ "step": 22
150
+ },
151
+ {
152
+ "epoch": 0.07,
153
+ "learning_rate": 0.00019991526325722315,
154
+ "loss": 0.8219,
155
+ "step": 23
156
+ },
157
+ {
158
+ "epoch": 0.07,
159
+ "learning_rate": 0.00019990172765214128,
160
+ "loss": 0.7624,
161
+ "step": 24
162
+ },
163
+ {
164
+ "epoch": 0.07,
165
+ "learning_rate": 0.00019988719009016753,
166
+ "loss": 0.7527,
167
+ "step": 25
168
+ },
169
+ {
170
+ "epoch": 0.08,
171
+ "learning_rate": 0.00019987165071710527,
172
+ "loss": 0.7699,
173
+ "step": 26
174
+ },
175
+ {
176
+ "epoch": 0.08,
177
+ "learning_rate": 0.00019985510968880557,
178
+ "loss": 0.7536,
179
+ "step": 27
180
+ },
181
+ {
182
+ "epoch": 0.08,
183
+ "learning_rate": 0.00019983756717116536,
184
+ "loss": 0.7863,
185
+ "step": 28
186
+ },
187
+ {
188
+ "epoch": 0.09,
189
+ "learning_rate": 0.000199819023340126,
190
+ "loss": 0.7166,
191
+ "step": 29
192
+ },
193
+ {
194
+ "epoch": 0.09,
195
+ "learning_rate": 0.0001997994783816715,
196
+ "loss": 0.7459,
197
+ "step": 30
198
+ },
199
+ {
200
+ "epoch": 0.09,
201
+ "learning_rate": 0.00019977893249182652,
202
+ "loss": 0.7376,
203
+ "step": 31
204
+ },
205
+ {
206
+ "epoch": 0.1,
207
+ "learning_rate": 0.00019975738587665456,
208
+ "loss": 0.7624,
209
+ "step": 32
210
+ },
211
+ {
212
+ "epoch": 0.1,
213
+ "learning_rate": 0.00019973483875225574,
214
+ "loss": 0.7329,
215
+ "step": 33
216
+ },
217
+ {
218
+ "epoch": 0.1,
219
+ "learning_rate": 0.00019971129134476473,
220
+ "loss": 0.7512,
221
+ "step": 34
222
+ },
223
+ {
224
+ "epoch": 0.1,
225
+ "learning_rate": 0.0001996867438903486,
226
+ "loss": 0.7211,
227
+ "step": 35
228
+ },
229
+ {
230
+ "epoch": 0.11,
231
+ "learning_rate": 0.00019966119663520412,
232
+ "loss": 0.7467,
233
+ "step": 36
234
+ },
235
+ {
236
+ "epoch": 0.11,
237
+ "learning_rate": 0.00019963464983555556,
238
+ "loss": 0.7186,
239
+ "step": 37
240
+ },
241
+ {
242
+ "epoch": 0.11,
243
+ "learning_rate": 0.0001996071037576521,
244
+ "loss": 0.7244,
245
+ "step": 38
246
+ },
247
+ {
248
+ "epoch": 0.12,
249
+ "learning_rate": 0.00019957855867776508,
250
+ "loss": 0.7071,
251
+ "step": 39
252
+ },
253
+ {
254
+ "epoch": 0.12,
255
+ "learning_rate": 0.00019954901488218515,
256
+ "loss": 0.7681,
257
+ "step": 40
258
+ },
259
+ {
260
+ "epoch": 0.12,
261
+ "eval_loss": 1.2323185205459595,
262
+ "eval_runtime": 15.7174,
263
+ "eval_samples_per_second": 37.093,
264
+ "eval_steps_per_second": 18.578,
265
+ "step": 40
266
+ },
267
+ {
268
+ "epoch": 0.12,
269
+ "learning_rate": 0.0001995184726672197,
270
+ "loss": 0.708,
271
+ "step": 41
272
+ },
273
+ {
274
+ "epoch": 0.13,
275
+ "learning_rate": 0.00019948693233918952,
276
+ "loss": 0.7443,
277
+ "step": 42
278
+ },
279
+ {
280
+ "epoch": 0.13,
281
+ "learning_rate": 0.000199454394214426,
282
+ "loss": 0.7079,
283
+ "step": 43
284
+ },
285
+ {
286
+ "epoch": 0.13,
287
+ "learning_rate": 0.0001994208586192678,
288
+ "loss": 0.6812,
289
+ "step": 44
290
+ },
291
+ {
292
+ "epoch": 0.13,
293
+ "learning_rate": 0.0001993863258900577,
294
+ "loss": 0.6792,
295
+ "step": 45
296
+ },
297
+ {
298
+ "epoch": 0.14,
299
+ "learning_rate": 0.00019935079637313906,
300
+ "loss": 0.6553,
301
+ "step": 46
302
+ },
303
+ {
304
+ "epoch": 0.14,
305
+ "learning_rate": 0.0001993142704248525,
306
+ "loss": 0.6936,
307
+ "step": 47
308
+ },
309
+ {
310
+ "epoch": 0.14,
311
+ "learning_rate": 0.00019927674841153237,
312
+ "loss": 0.74,
313
+ "step": 48
314
+ },
315
+ {
316
+ "epoch": 0.15,
317
+ "learning_rate": 0.00019923823070950282,
318
+ "loss": 0.7154,
319
+ "step": 49
320
+ },
321
+ {
322
+ "epoch": 0.15,
323
+ "learning_rate": 0.0001991987177050743,
324
+ "loss": 0.6902,
325
+ "step": 50
326
+ },
327
+ {
328
+ "epoch": 0.15,
329
+ "learning_rate": 0.0001991582097945395,
330
+ "loss": 0.7009,
331
+ "step": 51
332
+ },
333
+ {
334
+ "epoch": 0.16,
335
+ "learning_rate": 0.00019911670738416947,
336
+ "loss": 0.699,
337
+ "step": 52
338
+ },
339
+ {
340
+ "epoch": 0.16,
341
+ "learning_rate": 0.00019907421089020955,
342
+ "loss": 0.681,
343
+ "step": 53
344
+ },
345
+ {
346
+ "epoch": 0.16,
347
+ "learning_rate": 0.00019903072073887507,
348
+ "loss": 0.7097,
349
+ "step": 54
350
+ },
351
+ {
352
+ "epoch": 0.16,
353
+ "learning_rate": 0.00019898623736634737,
354
+ "loss": 0.6861,
355
+ "step": 55
356
+ },
357
+ {
358
+ "epoch": 0.17,
359
+ "learning_rate": 0.000198940761218769,
360
+ "loss": 0.7279,
361
+ "step": 56
362
+ },
363
+ {
364
+ "epoch": 0.17,
365
+ "learning_rate": 0.00019889429275223957,
366
+ "loss": 0.6784,
367
+ "step": 57
368
+ },
369
+ {
370
+ "epoch": 0.17,
371
+ "learning_rate": 0.00019884683243281116,
372
+ "loss": 0.7,
373
+ "step": 58
374
+ },
375
+ {
376
+ "epoch": 0.18,
377
+ "learning_rate": 0.00019879838073648341,
378
+ "loss": 0.6634,
379
+ "step": 59
380
+ },
381
+ {
382
+ "epoch": 0.18,
383
+ "learning_rate": 0.00019874893814919906,
384
+ "loss": 0.6647,
385
+ "step": 60
386
+ },
387
+ {
388
+ "epoch": 0.18,
389
+ "eval_loss": 1.1306297779083252,
390
+ "eval_runtime": 15.73,
391
+ "eval_samples_per_second": 37.063,
392
+ "eval_steps_per_second": 18.563,
393
+ "step": 60
394
+ },
395
+ {
396
+ "epoch": 0.18,
397
+ "learning_rate": 0.00019869850516683875,
398
+ "loss": 0.7377,
399
+ "step": 61
400
+ },
401
+ {
402
+ "epoch": 0.19,
403
+ "learning_rate": 0.00019864708229521636,
404
+ "loss": 0.6794,
405
+ "step": 62
406
+ },
407
+ {
408
+ "epoch": 0.19,
409
+ "learning_rate": 0.00019859467005007368,
410
+ "loss": 0.6744,
411
+ "step": 63
412
+ },
413
+ {
414
+ "epoch": 0.19,
415
+ "learning_rate": 0.0001985412689570754,
416
+ "loss": 0.6921,
417
+ "step": 64
418
+ },
419
+ {
420
+ "epoch": 0.19,
421
+ "learning_rate": 0.00019848687955180386,
422
+ "loss": 0.7389,
423
+ "step": 65
424
+ },
425
+ {
426
+ "epoch": 0.2,
427
+ "learning_rate": 0.00019843150237975344,
428
+ "loss": 0.6795,
429
+ "step": 66
430
+ },
431
+ {
432
+ "epoch": 0.2,
433
+ "learning_rate": 0.00019837513799632534,
434
+ "loss": 0.6917,
435
+ "step": 67
436
+ },
437
+ {
438
+ "epoch": 0.2,
439
+ "learning_rate": 0.00019831778696682194,
440
+ "loss": 0.66,
441
+ "step": 68
442
+ },
443
+ {
444
+ "epoch": 0.21,
445
+ "learning_rate": 0.00019825944986644114,
446
+ "loss": 0.6926,
447
+ "step": 69
448
+ },
449
+ {
450
+ "epoch": 0.21,
451
+ "learning_rate": 0.00019820012728027044,
452
+ "loss": 0.7265,
453
+ "step": 70
454
+ },
455
+ {
456
+ "epoch": 0.21,
457
+ "learning_rate": 0.00019813981980328129,
458
+ "loss": 0.733,
459
+ "step": 71
460
+ },
461
+ {
462
+ "epoch": 0.22,
463
+ "learning_rate": 0.00019807852804032305,
464
+ "loss": 0.6543,
465
+ "step": 72
466
+ },
467
+ {
468
+ "epoch": 0.22,
469
+ "learning_rate": 0.00019801625260611685,
470
+ "loss": 0.673,
471
+ "step": 73
472
+ },
473
+ {
474
+ "epoch": 0.22,
475
+ "learning_rate": 0.00019795299412524945,
476
+ "loss": 0.6773,
477
+ "step": 74
478
+ },
479
+ {
480
+ "epoch": 0.22,
481
+ "learning_rate": 0.0001978887532321671,
482
+ "loss": 0.6469,
483
+ "step": 75
484
+ },
485
+ {
486
+ "epoch": 0.23,
487
+ "learning_rate": 0.000197823530571169,
488
+ "loss": 0.7272,
489
+ "step": 76
490
+ },
491
+ {
492
+ "epoch": 0.23,
493
+ "learning_rate": 0.00019775732679640093,
494
+ "loss": 0.7015,
495
+ "step": 77
496
+ },
497
+ {
498
+ "epoch": 0.23,
499
+ "learning_rate": 0.0001976901425718487,
500
+ "loss": 0.6973,
501
+ "step": 78
502
+ },
503
+ {
504
+ "epoch": 0.24,
505
+ "learning_rate": 0.00019762197857133148,
506
+ "loss": 0.6831,
507
+ "step": 79
508
+ },
509
+ {
510
+ "epoch": 0.24,
511
+ "learning_rate": 0.00019755283547849494,
512
+ "loss": 0.6742,
513
+ "step": 80
514
+ },
515
+ {
516
+ "epoch": 0.24,
517
+ "eval_loss": 1.0846511125564575,
518
+ "eval_runtime": 15.6885,
519
+ "eval_samples_per_second": 37.161,
520
+ "eval_steps_per_second": 18.612,
521
+ "step": 80
522
+ },
523
+ {
524
+ "epoch": 0.24,
525
+ "learning_rate": 0.0001974827139868046,
526
+ "loss": 0.683,
527
+ "step": 81
528
+ },
529
+ {
530
+ "epoch": 0.24,
531
+ "learning_rate": 0.0001974116147995387,
532
+ "loss": 0.7063,
533
+ "step": 82
534
+ },
535
+ {
536
+ "epoch": 0.25,
537
+ "learning_rate": 0.00019733953862978123,
538
+ "loss": 0.6963,
539
+ "step": 83
540
+ },
541
+ {
542
+ "epoch": 0.25,
543
+ "learning_rate": 0.00019726648620041468,
544
+ "loss": 0.7179,
545
+ "step": 84
546
+ },
547
+ {
548
+ "epoch": 0.25,
549
+ "learning_rate": 0.00019719245824411296,
550
+ "loss": 0.7315,
551
+ "step": 85
552
+ },
553
+ {
554
+ "epoch": 0.26,
555
+ "learning_rate": 0.0001971174555033339,
556
+ "loss": 0.6839,
557
+ "step": 86
558
+ },
559
+ {
560
+ "epoch": 0.26,
561
+ "learning_rate": 0.00019704147873031186,
562
+ "loss": 0.6917,
563
+ "step": 87
564
+ },
565
+ {
566
+ "epoch": 0.26,
567
+ "learning_rate": 0.00019696452868705024,
568
+ "loss": 0.6828,
569
+ "step": 88
570
+ },
571
+ {
572
+ "epoch": 0.27,
573
+ "learning_rate": 0.00019688660614531367,
574
+ "loss": 0.6965,
575
+ "step": 89
576
+ },
577
+ {
578
+ "epoch": 0.27,
579
+ "learning_rate": 0.00019680771188662044,
580
+ "loss": 0.6522,
581
+ "step": 90
582
+ },
583
+ {
584
+ "epoch": 0.27,
585
+ "learning_rate": 0.00019672784670223464,
586
+ "loss": 0.6939,
587
+ "step": 91
588
+ },
589
+ {
590
+ "epoch": 0.27,
591
+ "learning_rate": 0.0001966470113931582,
592
+ "loss": 0.7037,
593
+ "step": 92
594
+ },
595
+ {
596
+ "epoch": 0.28,
597
+ "learning_rate": 0.00019656520677012264,
598
+ "loss": 0.6903,
599
+ "step": 93
600
+ },
601
+ {
602
+ "epoch": 0.28,
603
+ "learning_rate": 0.00019648243365358146,
604
+ "loss": 0.706,
605
+ "step": 94
606
+ },
607
+ {
608
+ "epoch": 0.28,
609
+ "learning_rate": 0.0001963986928737014,
610
+ "loss": 0.6253,
611
+ "step": 95
612
+ },
613
+ {
614
+ "epoch": 0.29,
615
+ "learning_rate": 0.00019631398527035422,
616
+ "loss": 0.732,
617
+ "step": 96
618
+ },
619
+ {
620
+ "epoch": 0.29,
621
+ "learning_rate": 0.0001962283116931086,
622
+ "loss": 0.6935,
623
+ "step": 97
624
+ },
625
+ {
626
+ "epoch": 0.29,
627
+ "learning_rate": 0.00019614167300122126,
628
+ "loss": 0.731,
629
+ "step": 98
630
+ },
631
+ {
632
+ "epoch": 0.3,
633
+ "learning_rate": 0.0001960540700636284,
634
+ "loss": 0.6916,
635
+ "step": 99
636
+ },
637
+ {
638
+ "epoch": 0.3,
639
+ "learning_rate": 0.0001959655037589372,
640
+ "loss": 0.6995,
641
+ "step": 100
642
+ },
643
+ {
644
+ "epoch": 0.3,
645
+ "eval_loss": 1.0572510957717896,
646
+ "eval_runtime": 15.7086,
647
+ "eval_samples_per_second": 37.113,
648
+ "eval_steps_per_second": 18.588,
649
+ "step": 100
650
+ },
651
+ {
652
+ "epoch": 0.3,
653
+ "learning_rate": 0.0001958759749754167,
654
+ "loss": 0.7313,
655
+ "step": 101
656
+ },
657
+ {
658
+ "epoch": 0.3,
659
+ "learning_rate": 0.00019578548461098914,
660
+ "loss": 0.7337,
661
+ "step": 102
662
+ },
663
+ {
664
+ "epoch": 0.31,
665
+ "learning_rate": 0.0001956940335732209,
666
+ "loss": 0.6721,
667
+ "step": 103
668
+ },
669
+ {
670
+ "epoch": 0.31,
671
+ "learning_rate": 0.00019560162277931325,
672
+ "loss": 0.6907,
673
+ "step": 104
674
+ },
675
+ {
676
+ "epoch": 0.31,
677
+ "learning_rate": 0.0001955082531560934,
678
+ "loss": 0.6599,
679
+ "step": 105
680
+ },
681
+ {
682
+ "epoch": 0.32,
683
+ "learning_rate": 0.00019541392564000488,
684
+ "loss": 0.6941,
685
+ "step": 106
686
+ },
687
+ {
688
+ "epoch": 0.32,
689
+ "learning_rate": 0.00019531864117709855,
690
+ "loss": 0.7037,
691
+ "step": 107
692
+ },
693
+ {
694
+ "epoch": 0.32,
695
+ "learning_rate": 0.00019522240072302274,
696
+ "loss": 0.6819,
697
+ "step": 108
698
+ },
699
+ {
700
+ "epoch": 0.33,
701
+ "learning_rate": 0.00019512520524301387,
702
+ "loss": 0.6729,
703
+ "step": 109
704
+ },
705
+ {
706
+ "epoch": 0.33,
707
+ "learning_rate": 0.00019502705571188672,
708
+ "loss": 0.713,
709
+ "step": 110
710
+ },
711
+ {
712
+ "epoch": 0.33,
713
+ "learning_rate": 0.00019492795311402473,
714
+ "loss": 0.6638,
715
+ "step": 111
716
+ },
717
+ {
718
+ "epoch": 0.33,
719
+ "learning_rate": 0.0001948278984433699,
720
+ "loss": 0.647,
721
+ "step": 112
722
+ },
723
+ {
724
+ "epoch": 0.34,
725
+ "learning_rate": 0.00019472689270341303,
726
+ "loss": 0.614,
727
+ "step": 113
728
+ },
729
+ {
730
+ "epoch": 0.34,
731
+ "learning_rate": 0.0001946249369071837,
732
+ "loss": 0.6783,
733
+ "step": 114
734
+ },
735
+ {
736
+ "epoch": 0.34,
737
+ "learning_rate": 0.0001945220320772399,
738
+ "loss": 0.7093,
739
+ "step": 115
740
+ },
741
+ {
742
+ "epoch": 0.35,
743
+ "learning_rate": 0.00019441817924565786,
744
+ "loss": 0.7179,
745
+ "step": 116
746
+ },
747
+ {
748
+ "epoch": 0.35,
749
+ "learning_rate": 0.00019431337945402185,
750
+ "loss": 0.7241,
751
+ "step": 117
752
+ },
753
+ {
754
+ "epoch": 0.35,
755
+ "learning_rate": 0.0001942076337534135,
756
+ "loss": 0.7168,
757
+ "step": 118
758
+ },
759
+ {
760
+ "epoch": 0.36,
761
+ "learning_rate": 0.00019410094320440138,
762
+ "loss": 0.7127,
763
+ "step": 119
764
+ },
765
+ {
766
+ "epoch": 0.36,
767
+ "learning_rate": 0.00019399330887703037,
768
+ "loss": 0.6883,
769
+ "step": 120
770
+ },
771
+ {
772
+ "epoch": 0.36,
773
+ "eval_loss": 1.0412259101867676,
774
+ "eval_runtime": 15.6448,
775
+ "eval_samples_per_second": 37.265,
776
+ "eval_steps_per_second": 18.664,
777
+ "step": 120
778
+ },
779
+ {
780
+ "epoch": 0.36,
781
+ "learning_rate": 0.00019388473185081087,
782
+ "loss": 0.6663,
783
+ "step": 121
784
+ },
785
+ {
786
+ "epoch": 0.36,
787
+ "learning_rate": 0.00019377521321470805,
788
+ "loss": 0.7319,
789
+ "step": 122
790
+ },
791
+ {
792
+ "epoch": 0.37,
793
+ "learning_rate": 0.00019366475406713086,
794
+ "loss": 0.6942,
795
+ "step": 123
796
+ },
797
+ {
798
+ "epoch": 0.37,
799
+ "learning_rate": 0.00019355335551592105,
800
+ "loss": 0.6847,
801
+ "step": 124
802
+ },
803
+ {
804
+ "epoch": 0.37,
805
+ "learning_rate": 0.000193441018678342,
806
+ "loss": 0.6732,
807
+ "step": 125
808
+ },
809
+ {
810
+ "epoch": 0.38,
811
+ "learning_rate": 0.00019332774468106768,
812
+ "loss": 0.6661,
813
+ "step": 126
814
+ },
815
+ {
816
+ "epoch": 0.38,
817
+ "learning_rate": 0.00019321353466017108,
818
+ "loss": 0.7044,
819
+ "step": 127
820
+ },
821
+ {
822
+ "epoch": 0.38,
823
+ "learning_rate": 0.00019309838976111311,
824
+ "loss": 0.6794,
825
+ "step": 128
826
+ },
827
+ {
828
+ "epoch": 0.39,
829
+ "learning_rate": 0.00019298231113873087,
830
+ "loss": 0.7275,
831
+ "step": 129
832
+ },
833
+ {
834
+ "epoch": 0.39,
835
+ "learning_rate": 0.00019286529995722623,
836
+ "loss": 0.685,
837
+ "step": 130
838
+ },
839
+ {
840
+ "epoch": 0.39,
841
+ "learning_rate": 0.00019274735739015404,
842
+ "loss": 0.6825,
843
+ "step": 131
844
+ },
845
+ {
846
+ "epoch": 0.39,
847
+ "learning_rate": 0.00019262848462041045,
848
+ "loss": 0.6857,
849
+ "step": 132
850
+ },
851
+ {
852
+ "epoch": 0.4,
853
+ "learning_rate": 0.000192508682840221,
854
+ "loss": 0.7139,
855
+ "step": 133
856
+ },
857
+ {
858
+ "epoch": 0.4,
859
+ "learning_rate": 0.0001923879532511287,
860
+ "loss": 0.6413,
861
+ "step": 134
862
+ },
863
+ {
864
+ "epoch": 0.4,
865
+ "learning_rate": 0.00019226629706398186,
866
+ "loss": 0.7257,
867
+ "step": 135
868
+ },
869
+ {
870
+ "epoch": 0.41,
871
+ "learning_rate": 0.0001921437154989221,
872
+ "loss": 0.6365,
873
+ "step": 136
874
+ },
875
+ {
876
+ "epoch": 0.41,
877
+ "learning_rate": 0.00019202020978537207,
878
+ "loss": 0.66,
879
+ "step": 137
880
+ },
881
+ {
882
+ "epoch": 0.41,
883
+ "learning_rate": 0.00019189578116202307,
884
+ "loss": 0.6525,
885
+ "step": 138
886
+ },
887
+ {
888
+ "epoch": 0.42,
889
+ "learning_rate": 0.00019177043087682265,
890
+ "loss": 0.6438,
891
+ "step": 139
892
+ },
893
+ {
894
+ "epoch": 0.42,
895
+ "learning_rate": 0.00019164416018696207,
896
+ "loss": 0.6437,
897
+ "step": 140
898
+ },
899
+ {
900
+ "epoch": 0.42,
901
+ "eval_loss": 1.0374538898468018,
902
+ "eval_runtime": 15.6903,
903
+ "eval_samples_per_second": 37.157,
904
+ "eval_steps_per_second": 18.61,
905
+ "step": 140
906
+ },
907
+ {
908
+ "epoch": 0.42,
909
+ "learning_rate": 0.00019151697035886382,
910
+ "loss": 0.6878,
911
+ "step": 141
912
+ },
913
+ {
914
+ "epoch": 0.42,
915
+ "learning_rate": 0.00019138886266816866,
916
+ "loss": 0.7072,
917
+ "step": 142
918
+ },
919
+ {
920
+ "epoch": 0.43,
921
+ "learning_rate": 0.00019125983839972317,
922
+ "loss": 0.69,
923
+ "step": 143
924
+ },
925
+ {
926
+ "epoch": 0.43,
927
+ "learning_rate": 0.00019112989884756653,
928
+ "loss": 0.7666,
929
+ "step": 144
930
+ },
931
+ {
932
+ "epoch": 0.43,
933
+ "learning_rate": 0.00019099904531491774,
934
+ "loss": 0.689,
935
+ "step": 145
936
+ },
937
+ {
938
+ "epoch": 0.44,
939
+ "learning_rate": 0.0001908672791141625,
940
+ "loss": 0.7394,
941
+ "step": 146
942
+ },
943
+ {
944
+ "epoch": 0.44,
945
+ "learning_rate": 0.0001907346015668401,
946
+ "loss": 0.6882,
947
+ "step": 147
948
+ },
949
+ {
950
+ "epoch": 0.44,
951
+ "learning_rate": 0.00019060101400362998,
952
+ "loss": 0.6394,
953
+ "step": 148
954
+ },
955
+ {
956
+ "epoch": 0.45,
957
+ "learning_rate": 0.00019046651776433862,
958
+ "loss": 0.6932,
959
+ "step": 149
960
+ },
961
+ {
962
+ "epoch": 0.45,
963
+ "learning_rate": 0.00019033111419788597,
964
+ "loss": 0.6676,
965
+ "step": 150
966
+ },
967
+ {
968
+ "epoch": 0.45,
969
+ "learning_rate": 0.00019019480466229201,
970
+ "loss": 0.6818,
971
+ "step": 151
972
+ },
973
+ {
974
+ "epoch": 0.45,
975
+ "learning_rate": 0.000190057590524663,
976
+ "loss": 0.6695,
977
+ "step": 152
978
+ },
979
+ {
980
+ "epoch": 0.46,
981
+ "learning_rate": 0.0001899194731611778,
982
+ "loss": 0.7082,
983
+ "step": 153
984
+ },
985
+ {
986
+ "epoch": 0.46,
987
+ "learning_rate": 0.00018978045395707418,
988
+ "loss": 0.6626,
989
+ "step": 154
990
+ },
991
+ {
992
+ "epoch": 0.46,
993
+ "learning_rate": 0.00018964053430663483,
994
+ "loss": 0.6708,
995
+ "step": 155
996
+ },
997
+ {
998
+ "epoch": 0.47,
999
+ "learning_rate": 0.0001894997156131734,
1000
+ "loss": 0.7864,
1001
+ "step": 156
1002
+ },
1003
+ {
1004
+ "epoch": 0.47,
1005
+ "learning_rate": 0.00018935799928902045,
1006
+ "loss": 0.6899,
1007
+ "step": 157
1008
+ },
1009
+ {
1010
+ "epoch": 0.47,
1011
+ "learning_rate": 0.0001892153867555092,
1012
+ "loss": 0.6574,
1013
+ "step": 158
1014
+ },
1015
+ {
1016
+ "epoch": 0.47,
1017
+ "learning_rate": 0.00018907187944296137,
1018
+ "loss": 0.7157,
1019
+ "step": 159
1020
+ },
1021
+ {
1022
+ "epoch": 0.48,
1023
+ "learning_rate": 0.00018892747879067286,
1024
+ "loss": 0.6331,
1025
+ "step": 160
1026
+ },
1027
+ {
1028
+ "epoch": 0.48,
1029
+ "eval_loss": 1.0186299085617065,
1030
+ "eval_runtime": 15.6809,
1031
+ "eval_samples_per_second": 37.179,
1032
+ "eval_steps_per_second": 18.621,
1033
+ "step": 160
1034
+ },
1035
+ {
1036
+ "epoch": 0.48,
1037
+ "learning_rate": 0.0001887821862468991,
1038
+ "loss": 0.7219,
1039
+ "step": 161
1040
+ },
1041
+ {
1042
+ "epoch": 0.48,
1043
+ "learning_rate": 0.00018863600326884082,
1044
+ "loss": 0.7203,
1045
+ "step": 162
1046
+ },
1047
+ {
1048
+ "epoch": 0.49,
1049
+ "learning_rate": 0.00018848893132262926,
1050
+ "loss": 0.6722,
1051
+ "step": 163
1052
+ },
1053
+ {
1054
+ "epoch": 0.49,
1055
+ "learning_rate": 0.00018834097188331143,
1056
+ "loss": 0.7223,
1057
+ "step": 164
1058
+ },
1059
+ {
1060
+ "epoch": 0.49,
1061
+ "learning_rate": 0.0001881921264348355,
1062
+ "loss": 0.659,
1063
+ "step": 165
1064
+ },
1065
+ {
1066
+ "epoch": 0.5,
1067
+ "learning_rate": 0.00018804239647003573,
1068
+ "loss": 0.6763,
1069
+ "step": 166
1070
+ },
1071
+ {
1072
+ "epoch": 0.5,
1073
+ "learning_rate": 0.00018789178349061753,
1074
+ "loss": 0.649,
1075
+ "step": 167
1076
+ },
1077
+ {
1078
+ "epoch": 0.5,
1079
+ "learning_rate": 0.00018774028900714256,
1080
+ "loss": 0.6762,
1081
+ "step": 168
1082
+ },
1083
+ {
1084
+ "epoch": 0.5,
1085
+ "learning_rate": 0.00018758791453901336,
1086
+ "loss": 0.6547,
1087
+ "step": 169
1088
+ },
1089
+ {
1090
+ "epoch": 0.51,
1091
+ "learning_rate": 0.00018743466161445823,
1092
+ "loss": 0.6516,
1093
+ "step": 170
1094
+ },
1095
+ {
1096
+ "epoch": 0.51,
1097
+ "learning_rate": 0.0001872805317705159,
1098
+ "loss": 0.6479,
1099
+ "step": 171
1100
+ },
1101
+ {
1102
+ "epoch": 0.51,
1103
+ "learning_rate": 0.0001871255265530201,
1104
+ "loss": 0.6986,
1105
+ "step": 172
1106
+ },
1107
+ {
1108
+ "epoch": 0.52,
1109
+ "learning_rate": 0.00018696964751658404,
1110
+ "loss": 0.7402,
1111
+ "step": 173
1112
+ },
1113
+ {
1114
+ "epoch": 0.52,
1115
+ "learning_rate": 0.00018681289622458485,
1116
+ "loss": 0.6629,
1117
+ "step": 174
1118
+ },
1119
+ {
1120
+ "epoch": 0.52,
1121
+ "learning_rate": 0.00018665527424914787,
1122
+ "loss": 0.6761,
1123
+ "step": 175
1124
+ },
1125
+ {
1126
+ "epoch": 0.53,
1127
+ "learning_rate": 0.00018649678317113084,
1128
+ "loss": 0.6746,
1129
+ "step": 176
1130
+ },
1131
+ {
1132
+ "epoch": 0.53,
1133
+ "learning_rate": 0.0001863374245801082,
1134
+ "loss": 0.67,
1135
+ "step": 177
1136
+ },
1137
+ {
1138
+ "epoch": 0.53,
1139
+ "learning_rate": 0.00018617720007435497,
1140
+ "loss": 0.7406,
1141
+ "step": 178
1142
+ },
1143
+ {
1144
+ "epoch": 0.53,
1145
+ "learning_rate": 0.00018601611126083087,
1146
+ "loss": 0.6735,
1147
+ "step": 179
1148
+ },
1149
+ {
1150
+ "epoch": 0.54,
1151
+ "learning_rate": 0.000185854159755164,
1152
+ "loss": 0.6686,
1153
+ "step": 180
1154
+ },
1155
+ {
1156
+ "epoch": 0.54,
1157
+ "eval_loss": 1.0153461694717407,
1158
+ "eval_runtime": 15.6769,
1159
+ "eval_samples_per_second": 37.189,
1160
+ "eval_steps_per_second": 18.626,
1161
+ "step": 180
1162
+ },
1163
+ {
1164
+ "epoch": 0.54,
1165
+ "learning_rate": 0.00018569134718163496,
1166
+ "loss": 0.6416,
1167
+ "step": 181
1168
+ },
1169
+ {
1170
+ "epoch": 0.54,
1171
+ "learning_rate": 0.00018552767517316022,
1172
+ "loss": 0.6456,
1173
+ "step": 182
1174
+ },
1175
+ {
1176
+ "epoch": 0.55,
1177
+ "learning_rate": 0.00018536314537127586,
1178
+ "loss": 0.6794,
1179
+ "step": 183
1180
+ },
1181
+ {
1182
+ "epoch": 0.55,
1183
+ "learning_rate": 0.00018519775942612128,
1184
+ "loss": 0.6754,
1185
+ "step": 184
1186
+ },
1187
+ {
1188
+ "epoch": 0.55,
1189
+ "learning_rate": 0.00018503151899642236,
1190
+ "loss": 0.6251,
1191
+ "step": 185
1192
+ },
1193
+ {
1194
+ "epoch": 0.56,
1195
+ "learning_rate": 0.00018486442574947511,
1196
+ "loss": 0.7354,
1197
+ "step": 186
1198
+ },
1199
+ {
1200
+ "epoch": 0.56,
1201
+ "learning_rate": 0.00018469648136112867,
1202
+ "loss": 0.6457,
1203
+ "step": 187
1204
+ },
1205
+ {
1206
+ "epoch": 0.56,
1207
+ "learning_rate": 0.0001845276875157687,
1208
+ "loss": 0.6823,
1209
+ "step": 188
1210
+ },
1211
+ {
1212
+ "epoch": 0.56,
1213
+ "learning_rate": 0.00018435804590630043,
1214
+ "loss": 0.6891,
1215
+ "step": 189
1216
+ },
1217
+ {
1218
+ "epoch": 0.57,
1219
+ "learning_rate": 0.0001841875582341317,
1220
+ "loss": 0.6731,
1221
+ "step": 190
1222
+ },
1223
+ {
1224
+ "epoch": 0.57,
1225
+ "learning_rate": 0.00018401622620915583,
1226
+ "loss": 0.6857,
1227
+ "step": 191
1228
+ },
1229
+ {
1230
+ "epoch": 0.57,
1231
+ "learning_rate": 0.0001838440515497345,
1232
+ "loss": 0.6807,
1233
+ "step": 192
1234
+ },
1235
+ {
1236
+ "epoch": 0.58,
1237
+ "learning_rate": 0.0001836710359826806,
1238
+ "loss": 0.6622,
1239
+ "step": 193
1240
+ },
1241
+ {
1242
+ "epoch": 0.58,
1243
+ "learning_rate": 0.00018349718124324076,
1244
+ "loss": 0.6628,
1245
+ "step": 194
1246
+ },
1247
+ {
1248
+ "epoch": 0.58,
1249
+ "learning_rate": 0.00018332248907507807,
1250
+ "loss": 0.724,
1251
+ "step": 195
1252
+ },
1253
+ {
1254
+ "epoch": 0.59,
1255
+ "learning_rate": 0.00018314696123025454,
1256
+ "loss": 0.675,
1257
+ "step": 196
1258
+ },
1259
+ {
1260
+ "epoch": 0.59,
1261
+ "learning_rate": 0.00018297059946921355,
1262
+ "loss": 0.677,
1263
+ "step": 197
1264
+ },
1265
+ {
1266
+ "epoch": 0.59,
1267
+ "learning_rate": 0.00018279340556076216,
1268
+ "loss": 0.6555,
1269
+ "step": 198
1270
+ },
1271
+ {
1272
+ "epoch": 0.59,
1273
+ "learning_rate": 0.0001826153812820534,
1274
+ "loss": 0.6273,
1275
+ "step": 199
1276
+ },
1277
+ {
1278
+ "epoch": 0.6,
1279
+ "learning_rate": 0.0001824365284185684,
1280
+ "loss": 0.6767,
1281
+ "step": 200
1282
+ },
1283
+ {
1284
+ "epoch": 0.6,
1285
+ "eval_loss": 1.004173994064331,
1286
+ "eval_runtime": 15.726,
1287
+ "eval_samples_per_second": 37.072,
1288
+ "eval_steps_per_second": 18.568,
1289
+ "step": 200
1290
+ },
1291
+ {
1292
+ "epoch": 0.6,
1293
+ "learning_rate": 0.00018225684876409863,
1294
+ "loss": 0.6719,
1295
+ "step": 201
1296
+ },
1297
+ {
1298
+ "epoch": 0.6,
1299
+ "learning_rate": 0.00018207634412072764,
1300
+ "loss": 0.6522,
1301
+ "step": 202
1302
+ },
1303
+ {
1304
+ "epoch": 0.61,
1305
+ "learning_rate": 0.0001818950162988133,
1306
+ "loss": 0.6957,
1307
+ "step": 203
1308
+ },
1309
+ {
1310
+ "epoch": 0.61,
1311
+ "learning_rate": 0.00018171286711696934,
1312
+ "loss": 0.7013,
1313
+ "step": 204
1314
+ },
1315
+ {
1316
+ "epoch": 0.61,
1317
+ "learning_rate": 0.00018152989840204735,
1318
+ "loss": 0.6603,
1319
+ "step": 205
1320
+ },
1321
+ {
1322
+ "epoch": 0.62,
1323
+ "learning_rate": 0.0001813461119891184,
1324
+ "loss": 0.6788,
1325
+ "step": 206
1326
+ },
1327
+ {
1328
+ "epoch": 0.62,
1329
+ "learning_rate": 0.0001811615097214545,
1330
+ "loss": 0.6795,
1331
+ "step": 207
1332
+ },
1333
+ {
1334
+ "epoch": 0.62,
1335
+ "learning_rate": 0.00018097609345051025,
1336
+ "loss": 0.6458,
1337
+ "step": 208
1338
+ },
1339
+ {
1340
+ "epoch": 0.62,
1341
+ "learning_rate": 0.00018078986503590428,
1342
+ "loss": 0.6741,
1343
+ "step": 209
1344
+ },
1345
+ {
1346
+ "epoch": 0.63,
1347
+ "learning_rate": 0.00018060282634540053,
1348
+ "loss": 0.6537,
1349
+ "step": 210
1350
+ },
1351
+ {
1352
+ "epoch": 0.63,
1353
+ "learning_rate": 0.0001804149792548895,
1354
+ "loss": 0.6501,
1355
+ "step": 211
1356
+ },
1357
+ {
1358
+ "epoch": 0.63,
1359
+ "learning_rate": 0.00018022632564836948,
1360
+ "loss": 0.673,
1361
+ "step": 212
1362
+ },
1363
+ {
1364
+ "epoch": 0.64,
1365
+ "learning_rate": 0.00018003686741792772,
1366
+ "loss": 0.6746,
1367
+ "step": 213
1368
+ },
1369
+ {
1370
+ "epoch": 0.64,
1371
+ "learning_rate": 0.0001798466064637214,
1372
+ "loss": 0.6329,
1373
+ "step": 214
1374
+ },
1375
+ {
1376
+ "epoch": 0.64,
1377
+ "learning_rate": 0.00017965554469395837,
1378
+ "loss": 0.69,
1379
+ "step": 215
1380
+ },
1381
+ {
1382
+ "epoch": 0.65,
1383
+ "learning_rate": 0.00017946368402487845,
1384
+ "loss": 0.6513,
1385
+ "step": 216
1386
+ },
1387
+ {
1388
+ "epoch": 0.65,
1389
+ "learning_rate": 0.00017927102638073383,
1390
+ "loss": 0.6758,
1391
+ "step": 217
1392
+ },
1393
+ {
1394
+ "epoch": 0.65,
1395
+ "learning_rate": 0.00017907757369376985,
1396
+ "loss": 0.6776,
1397
+ "step": 218
1398
+ },
1399
+ {
1400
+ "epoch": 0.65,
1401
+ "learning_rate": 0.00017888332790420584,
1402
+ "loss": 0.6487,
1403
+ "step": 219
1404
+ },
1405
+ {
1406
+ "epoch": 0.66,
1407
+ "learning_rate": 0.00017868829096021527,
1408
+ "loss": 0.7037,
1409
+ "step": 220
1410
+ },
1411
+ {
1412
+ "epoch": 0.66,
1413
+ "eval_loss": 1.002336025238037,
1414
+ "eval_runtime": 17.4124,
1415
+ "eval_samples_per_second": 33.482,
1416
+ "eval_steps_per_second": 16.77,
1417
+ "step": 220
1418
+ },
1419
+ {
1420
+ "epoch": 0.66,
1421
+ "learning_rate": 0.00017849246481790664,
1422
+ "loss": 0.71,
1423
+ "step": 221
1424
+ },
1425
+ {
1426
+ "epoch": 0.66,
1427
+ "learning_rate": 0.00017829585144130356,
1428
+ "loss": 0.6695,
1429
+ "step": 222
1430
+ },
1431
+ {
1432
+ "epoch": 0.67,
1433
+ "learning_rate": 0.00017809845280232514,
1434
+ "loss": 0.6645,
1435
+ "step": 223
1436
+ },
1437
+ {
1438
+ "epoch": 0.67,
1439
+ "learning_rate": 0.0001779002708807662,
1440
+ "loss": 0.6983,
1441
+ "step": 224
1442
+ },
1443
+ {
1444
+ "epoch": 0.67,
1445
+ "learning_rate": 0.0001777013076642776,
1446
+ "loss": 0.6785,
1447
+ "step": 225
1448
+ },
1449
+ {
1450
+ "epoch": 0.68,
1451
+ "learning_rate": 0.0001775015651483459,
1452
+ "loss": 0.6791,
1453
+ "step": 226
1454
+ },
1455
+ {
1456
+ "epoch": 0.68,
1457
+ "learning_rate": 0.0001773010453362737,
1458
+ "loss": 0.695,
1459
+ "step": 227
1460
+ },
1461
+ {
1462
+ "epoch": 0.68,
1463
+ "learning_rate": 0.00017709975023915949,
1464
+ "loss": 0.6352,
1465
+ "step": 228
1466
+ },
1467
+ {
1468
+ "epoch": 0.68,
1469
+ "learning_rate": 0.00017689768187587732,
1470
+ "loss": 0.6911,
1471
+ "step": 229
1472
+ },
1473
+ {
1474
+ "epoch": 0.69,
1475
+ "learning_rate": 0.0001766948422730567,
1476
+ "loss": 0.7189,
1477
+ "step": 230
1478
+ },
1479
+ {
1480
+ "epoch": 0.69,
1481
+ "learning_rate": 0.00017649123346506223,
1482
+ "loss": 0.6838,
1483
+ "step": 231
1484
+ },
1485
+ {
1486
+ "epoch": 0.69,
1487
+ "learning_rate": 0.0001762868574939732,
1488
+ "loss": 0.7209,
1489
+ "step": 232
1490
+ },
1491
+ {
1492
+ "epoch": 0.7,
1493
+ "learning_rate": 0.00017608171640956313,
1494
+ "loss": 0.6719,
1495
+ "step": 233
1496
+ },
1497
+ {
1498
+ "epoch": 0.7,
1499
+ "learning_rate": 0.0001758758122692791,
1500
+ "loss": 0.6624,
1501
+ "step": 234
1502
+ },
1503
+ {
1504
+ "epoch": 0.7,
1505
+ "learning_rate": 0.0001756691471382213,
1506
+ "loss": 0.637,
1507
+ "step": 235
1508
+ },
1509
+ {
1510
+ "epoch": 0.71,
1511
+ "learning_rate": 0.00017546172308912213,
1512
+ "loss": 0.653,
1513
+ "step": 236
1514
+ },
1515
+ {
1516
+ "epoch": 0.71,
1517
+ "learning_rate": 0.0001752535422023256,
1518
+ "loss": 0.6368,
1519
+ "step": 237
1520
+ },
1521
+ {
1522
+ "epoch": 0.71,
1523
+ "learning_rate": 0.00017504460656576627,
1524
+ "loss": 0.6713,
1525
+ "step": 238
1526
+ },
1527
+ {
1528
+ "epoch": 0.71,
1529
+ "learning_rate": 0.00017483491827494847,
1530
+ "loss": 0.6614,
1531
+ "step": 239
1532
+ },
1533
+ {
1534
+ "epoch": 0.72,
1535
+ "learning_rate": 0.0001746244794329252,
1536
+ "loss": 0.6994,
1537
+ "step": 240
1538
+ },
1539
+ {
1540
+ "epoch": 0.72,
1541
+ "eval_loss": 1.0014034509658813,
1542
+ "eval_runtime": 15.6575,
1543
+ "eval_samples_per_second": 37.235,
1544
+ "eval_steps_per_second": 18.649,
1545
+ "step": 240
1546
+ },
1547
+ {
1548
+ "epoch": 0.72,
1549
+ "learning_rate": 0.00017441329215027706,
1550
+ "loss": 0.6644,
1551
+ "step": 241
1552
+ },
1553
+ {
1554
+ "epoch": 0.72,
1555
+ "learning_rate": 0.0001742013585450911,
1556
+ "loss": 0.6374,
1557
+ "step": 242
1558
+ },
1559
+ {
1560
+ "epoch": 0.73,
1561
+ "learning_rate": 0.00017398868074293948,
1562
+ "loss": 0.6627,
1563
+ "step": 243
1564
+ },
1565
+ {
1566
+ "epoch": 0.73,
1567
+ "learning_rate": 0.00017377526087685832,
1568
+ "loss": 0.6936,
1569
+ "step": 244
1570
+ },
1571
+ {
1572
+ "epoch": 0.73,
1573
+ "learning_rate": 0.00017356110108732617,
1574
+ "loss": 0.6419,
1575
+ "step": 245
1576
+ },
1577
+ {
1578
+ "epoch": 0.73,
1579
+ "learning_rate": 0.0001733462035222426,
1580
+ "loss": 0.6402,
1581
+ "step": 246
1582
+ },
1583
+ {
1584
+ "epoch": 0.74,
1585
+ "learning_rate": 0.0001731305703369066,
1586
+ "loss": 0.7028,
1587
+ "step": 247
1588
+ },
1589
+ {
1590
+ "epoch": 0.74,
1591
+ "learning_rate": 0.0001729142036939951,
1592
+ "loss": 0.6146,
1593
+ "step": 248
1594
+ },
1595
+ {
1596
+ "epoch": 0.74,
1597
+ "learning_rate": 0.0001726971057635411,
1598
+ "loss": 0.6702,
1599
+ "step": 249
1600
+ },
1601
+ {
1602
+ "epoch": 0.75,
1603
+ "learning_rate": 0.000172479278722912,
1604
+ "loss": 0.6971,
1605
+ "step": 250
1606
+ },
1607
+ {
1608
+ "epoch": 0.75,
1609
+ "learning_rate": 0.00017226072475678782,
1610
+ "loss": 0.6289,
1611
+ "step": 251
1612
+ },
1613
+ {
1614
+ "epoch": 0.75,
1615
+ "learning_rate": 0.0001720414460571392,
1616
+ "loss": 0.6786,
1617
+ "step": 252
1618
+ },
1619
+ {
1620
+ "epoch": 0.76,
1621
+ "learning_rate": 0.0001718214448232054,
1622
+ "loss": 0.6599,
1623
+ "step": 253
1624
+ },
1625
+ {
1626
+ "epoch": 0.76,
1627
+ "learning_rate": 0.0001716007232614723,
1628
+ "loss": 0.656,
1629
+ "step": 254
1630
+ },
1631
+ {
1632
+ "epoch": 0.76,
1633
+ "learning_rate": 0.00017137928358565028,
1634
+ "loss": 0.7092,
1635
+ "step": 255
1636
+ },
1637
+ {
1638
+ "epoch": 0.76,
1639
+ "learning_rate": 0.000171157128016652,
1640
+ "loss": 0.6596,
1641
+ "step": 256
1642
+ },
1643
+ {
1644
+ "epoch": 0.77,
1645
+ "learning_rate": 0.00017093425878257005,
1646
+ "loss": 0.6937,
1647
+ "step": 257
1648
+ },
1649
+ {
1650
+ "epoch": 0.77,
1651
+ "learning_rate": 0.00017071067811865476,
1652
+ "loss": 0.7027,
1653
+ "step": 258
1654
+ },
1655
+ {
1656
+ "epoch": 0.77,
1657
+ "learning_rate": 0.00017048638826729163,
1658
+ "loss": 0.6776,
1659
+ "step": 259
1660
+ },
1661
+ {
1662
+ "epoch": 0.78,
1663
+ "learning_rate": 0.0001702613914779789,
1664
+ "loss": 0.7012,
1665
+ "step": 260
1666
+ },
1667
+ {
1668
+ "epoch": 0.78,
1669
+ "eval_loss": 0.9995876550674438,
1670
+ "eval_runtime": 15.664,
1671
+ "eval_samples_per_second": 37.219,
1672
+ "eval_steps_per_second": 18.641,
1673
+ "step": 260
1674
+ },
1675
+ {
1676
+ "epoch": 0.78,
1677
+ "learning_rate": 0.00017003569000730498,
1678
+ "loss": 0.6786,
1679
+ "step": 261
1680
+ },
1681
+ {
1682
+ "epoch": 0.78,
1683
+ "learning_rate": 0.0001698092861189259,
1684
+ "loss": 0.7071,
1685
+ "step": 262
1686
+ },
1687
+ {
1688
+ "epoch": 0.79,
1689
+ "learning_rate": 0.00016958218208354252,
1690
+ "loss": 0.6556,
1691
+ "step": 263
1692
+ },
1693
+ {
1694
+ "epoch": 0.79,
1695
+ "learning_rate": 0.00016935438017887772,
1696
+ "loss": 0.6418,
1697
+ "step": 264
1698
+ },
1699
+ {
1700
+ "epoch": 0.79,
1701
+ "learning_rate": 0.0001691258826896536,
1702
+ "loss": 0.6726,
1703
+ "step": 265
1704
+ },
1705
+ {
1706
+ "epoch": 0.79,
1707
+ "learning_rate": 0.00016889669190756868,
1708
+ "loss": 0.6514,
1709
+ "step": 266
1710
+ },
1711
+ {
1712
+ "epoch": 0.8,
1713
+ "learning_rate": 0.00016866681013127466,
1714
+ "loss": 0.6914,
1715
+ "step": 267
1716
+ },
1717
+ {
1718
+ "epoch": 0.8,
1719
+ "learning_rate": 0.00016843623966635366,
1720
+ "loss": 0.6884,
1721
+ "step": 268
1722
+ },
1723
+ {
1724
+ "epoch": 0.8,
1725
+ "learning_rate": 0.00016820498282529479,
1726
+ "loss": 0.6624,
1727
+ "step": 269
1728
+ },
1729
+ {
1730
+ "epoch": 0.81,
1731
+ "learning_rate": 0.0001679730419274713,
1732
+ "loss": 0.7241,
1733
+ "step": 270
1734
+ },
1735
+ {
1736
+ "epoch": 0.81,
1737
+ "learning_rate": 0.00016774041929911698,
1738
+ "loss": 0.6923,
1739
+ "step": 271
1740
+ },
1741
+ {
1742
+ "epoch": 0.81,
1743
+ "learning_rate": 0.0001675071172733031,
1744
+ "loss": 0.6817,
1745
+ "step": 272
1746
+ },
1747
+ {
1748
+ "epoch": 0.82,
1749
+ "learning_rate": 0.00016727313818991487,
1750
+ "loss": 0.7042,
1751
+ "step": 273
1752
+ },
1753
+ {
1754
+ "epoch": 0.82,
1755
+ "learning_rate": 0.00016703848439562785,
1756
+ "loss": 0.66,
1757
+ "step": 274
1758
+ },
1759
+ {
1760
+ "epoch": 0.82,
1761
+ "learning_rate": 0.00016680315824388483,
1762
+ "loss": 0.662,
1763
+ "step": 275
1764
+ },
1765
+ {
1766
+ "epoch": 0.82,
1767
+ "learning_rate": 0.00016656716209487174,
1768
+ "loss": 0.6174,
1769
+ "step": 276
1770
+ },
1771
+ {
1772
+ "epoch": 0.83,
1773
+ "learning_rate": 0.00016633049831549424,
1774
+ "loss": 0.6684,
1775
+ "step": 277
1776
+ },
1777
+ {
1778
+ "epoch": 0.83,
1779
+ "learning_rate": 0.0001660931692793541,
1780
+ "loss": 0.6581,
1781
+ "step": 278
1782
+ },
1783
+ {
1784
+ "epoch": 0.83,
1785
+ "learning_rate": 0.00016585517736672498,
1786
+ "loss": 0.6835,
1787
+ "step": 279
1788
+ },
1789
+ {
1790
+ "epoch": 0.84,
1791
+ "learning_rate": 0.000165616524964529,
1792
+ "loss": 0.6599,
1793
+ "step": 280
1794
+ },
1795
+ {
1796
+ "epoch": 0.84,
1797
+ "eval_loss": 0.9926242828369141,
1798
+ "eval_runtime": 15.669,
1799
+ "eval_samples_per_second": 37.207,
1800
+ "eval_steps_per_second": 18.636,
1801
+ "step": 280
1802
+ },
1803
+ {
1804
+ "epoch": 0.84,
1805
+ "learning_rate": 0.00016537721446631252,
1806
+ "loss": 0.6619,
1807
+ "step": 281
1808
+ },
1809
+ {
1810
+ "epoch": 0.84,
1811
+ "learning_rate": 0.00016513724827222227,
1812
+ "loss": 0.6557,
1813
+ "step": 282
1814
+ },
1815
+ {
1816
+ "epoch": 0.85,
1817
+ "learning_rate": 0.00016489662878898113,
1818
+ "loss": 0.6577,
1819
+ "step": 283
1820
+ },
1821
+ {
1822
+ "epoch": 0.85,
1823
+ "learning_rate": 0.00016465535842986434,
1824
+ "loss": 0.6423,
1825
+ "step": 284
1826
+ },
1827
+ {
1828
+ "epoch": 0.85,
1829
+ "learning_rate": 0.00016441343961467471,
1830
+ "loss": 0.6472,
1831
+ "step": 285
1832
+ },
1833
+ {
1834
+ "epoch": 0.85,
1835
+ "learning_rate": 0.000164170874769719,
1836
+ "loss": 0.6434,
1837
+ "step": 286
1838
+ },
1839
+ {
1840
+ "epoch": 0.86,
1841
+ "learning_rate": 0.00016392766632778312,
1842
+ "loss": 0.6461,
1843
+ "step": 287
1844
+ },
1845
+ {
1846
+ "epoch": 0.86,
1847
+ "learning_rate": 0.00016368381672810786,
1848
+ "loss": 0.6857,
1849
+ "step": 288
1850
+ },
1851
+ {
1852
+ "epoch": 0.86,
1853
+ "learning_rate": 0.00016343932841636456,
1854
+ "loss": 0.6927,
1855
+ "step": 289
1856
+ },
1857
+ {
1858
+ "epoch": 0.87,
1859
+ "learning_rate": 0.0001631942038446304,
1860
+ "loss": 0.6263,
1861
+ "step": 290
1862
+ },
1863
+ {
1864
+ "epoch": 0.87,
1865
+ "learning_rate": 0.00016294844547136395,
1866
+ "loss": 0.6857,
1867
+ "step": 291
1868
+ },
1869
+ {
1870
+ "epoch": 0.87,
1871
+ "learning_rate": 0.00016270205576138032,
1872
+ "loss": 0.682,
1873
+ "step": 292
1874
+ },
1875
+ {
1876
+ "epoch": 0.88,
1877
+ "learning_rate": 0.0001624550371858267,
1878
+ "loss": 0.6964,
1879
+ "step": 293
1880
+ },
1881
+ {
1882
+ "epoch": 0.88,
1883
+ "learning_rate": 0.00016220739222215738,
1884
+ "loss": 0.6261,
1885
+ "step": 294
1886
+ },
1887
+ {
1888
+ "epoch": 0.88,
1889
+ "learning_rate": 0.00016195912335410895,
1890
+ "loss": 0.6848,
1891
+ "step": 295
1892
+ },
1893
+ {
1894
+ "epoch": 0.88,
1895
+ "learning_rate": 0.00016171023307167545,
1896
+ "loss": 0.635,
1897
+ "step": 296
1898
+ },
1899
+ {
1900
+ "epoch": 0.89,
1901
+ "learning_rate": 0.0001614607238710833,
1902
+ "loss": 0.6941,
1903
+ "step": 297
1904
+ },
1905
+ {
1906
+ "epoch": 0.89,
1907
+ "learning_rate": 0.0001612105982547663,
1908
+ "loss": 0.6339,
1909
+ "step": 298
1910
+ },
1911
+ {
1912
+ "epoch": 0.89,
1913
+ "learning_rate": 0.0001609598587313406,
1914
+ "loss": 0.6298,
1915
+ "step": 299
1916
+ },
1917
+ {
1918
+ "epoch": 0.9,
1919
+ "learning_rate": 0.00016070850781557948,
1920
+ "loss": 0.6401,
1921
+ "step": 300
1922
+ },
1923
+ {
1924
+ "epoch": 0.9,
1925
+ "eval_loss": 0.9913042187690735,
1926
+ "eval_runtime": 15.6715,
1927
+ "eval_samples_per_second": 37.201,
1928
+ "eval_steps_per_second": 18.633,
1929
+ "step": 300
1930
+ },
1931
+ {
1932
+ "epoch": 0.9,
1933
+ "learning_rate": 0.00016045654802838812,
1934
+ "loss": 0.6483,
1935
+ "step": 301
1936
+ },
1937
+ {
1938
+ "epoch": 0.9,
1939
+ "learning_rate": 0.0001602039818967783,
1940
+ "loss": 0.6352,
1941
+ "step": 302
1942
+ },
1943
+ {
1944
+ "epoch": 0.91,
1945
+ "learning_rate": 0.00015995081195384315,
1946
+ "loss": 0.6344,
1947
+ "step": 303
1948
+ },
1949
+ {
1950
+ "epoch": 0.91,
1951
+ "learning_rate": 0.00015969704073873157,
1952
+ "loss": 0.6317,
1953
+ "step": 304
1954
+ },
1955
+ {
1956
+ "epoch": 0.91,
1957
+ "learning_rate": 0.000159442670796623,
1958
+ "loss": 0.6979,
1959
+ "step": 305
1960
+ },
1961
+ {
1962
+ "epoch": 0.91,
1963
+ "learning_rate": 0.0001591877046787017,
1964
+ "loss": 0.6429,
1965
+ "step": 306
1966
+ },
1967
+ {
1968
+ "epoch": 0.92,
1969
+ "learning_rate": 0.00015893214494213132,
1970
+ "loss": 0.6493,
1971
+ "step": 307
1972
+ },
1973
+ {
1974
+ "epoch": 0.92,
1975
+ "learning_rate": 0.00015867599415002895,
1976
+ "loss": 0.6765,
1977
+ "step": 308
1978
+ },
1979
+ {
1980
+ "epoch": 0.92,
1981
+ "learning_rate": 0.00015841925487143987,
1982
+ "loss": 0.6871,
1983
+ "step": 309
1984
+ },
1985
+ {
1986
+ "epoch": 0.93,
1987
+ "learning_rate": 0.00015816192968131138,
1988
+ "loss": 0.6389,
1989
+ "step": 310
1990
+ },
1991
+ {
1992
+ "epoch": 0.93,
1993
+ "learning_rate": 0.00015790402116046726,
1994
+ "loss": 0.6002,
1995
+ "step": 311
1996
+ },
1997
+ {
1998
+ "epoch": 0.93,
1999
+ "learning_rate": 0.0001576455318955816,
2000
+ "loss": 0.6025,
2001
+ "step": 312
2002
+ },
2003
+ {
2004
+ "epoch": 0.94,
2005
+ "learning_rate": 0.0001573864644791532,
2006
+ "loss": 0.6692,
2007
+ "step": 313
2008
+ },
2009
+ {
2010
+ "epoch": 0.94,
2011
+ "learning_rate": 0.00015712682150947923,
2012
+ "loss": 0.7079,
2013
+ "step": 314
2014
+ },
2015
+ {
2016
+ "epoch": 0.94,
2017
+ "learning_rate": 0.00015686660559062945,
2018
+ "loss": 0.6622,
2019
+ "step": 315
2020
+ },
2021
+ {
2022
+ "epoch": 0.94,
2023
+ "learning_rate": 0.00015660581933241993,
2024
+ "loss": 0.6402,
2025
+ "step": 316
2026
+ },
2027
+ {
2028
+ "epoch": 0.95,
2029
+ "learning_rate": 0.00015634446535038688,
2030
+ "loss": 0.7251,
2031
+ "step": 317
2032
+ },
2033
+ {
2034
+ "epoch": 0.95,
2035
+ "learning_rate": 0.00015608254626576048,
2036
+ "loss": 0.6806,
2037
+ "step": 318
2038
+ },
2039
+ {
2040
+ "epoch": 0.95,
2041
+ "learning_rate": 0.00015582006470543855,
2042
+ "loss": 0.6475,
2043
+ "step": 319
2044
+ },
2045
+ {
2046
+ "epoch": 0.96,
2047
+ "learning_rate": 0.00015555702330196023,
2048
+ "loss": 0.6665,
2049
+ "step": 320
2050
+ },
2051
+ {
2052
+ "epoch": 0.96,
2053
+ "eval_loss": 0.9910463094711304,
2054
+ "eval_runtime": 15.7019,
2055
+ "eval_samples_per_second": 37.129,
2056
+ "eval_steps_per_second": 18.596,
2057
+ "step": 320
2058
+ },
2059
+ {
2060
+ "epoch": 0.96,
2061
+ "learning_rate": 0.00015529342469347958,
2062
+ "loss": 0.6674,
2063
+ "step": 321
2064
+ },
2065
+ {
2066
+ "epoch": 0.96,
2067
+ "learning_rate": 0.00015502927152373914,
2068
+ "loss": 0.6873,
2069
+ "step": 322
2070
+ },
2071
+ {
2072
+ "epoch": 0.96,
2073
+ "learning_rate": 0.00015476456644204328,
2074
+ "loss": 0.6425,
2075
+ "step": 323
2076
+ },
2077
+ {
2078
+ "epoch": 0.97,
2079
+ "learning_rate": 0.0001544993121032318,
2080
+ "loss": 0.6565,
2081
+ "step": 324
2082
+ },
2083
+ {
2084
+ "epoch": 0.97,
2085
+ "learning_rate": 0.00015423351116765317,
2086
+ "loss": 0.6886,
2087
+ "step": 325
2088
+ },
2089
+ {
2090
+ "epoch": 0.97,
2091
+ "learning_rate": 0.000153967166301138,
2092
+ "loss": 0.6904,
2093
+ "step": 326
2094
+ },
2095
+ {
2096
+ "epoch": 0.98,
2097
+ "learning_rate": 0.00015370028017497217,
2098
+ "loss": 0.6281,
2099
+ "step": 327
2100
+ },
2101
+ {
2102
+ "epoch": 0.98,
2103
+ "learning_rate": 0.00015343285546587013,
2104
+ "loss": 0.6836,
2105
+ "step": 328
2106
+ },
2107
+ {
2108
+ "epoch": 0.98,
2109
+ "learning_rate": 0.00015316489485594782,
2110
+ "loss": 0.697,
2111
+ "step": 329
2112
+ },
2113
+ {
2114
+ "epoch": 0.99,
2115
+ "learning_rate": 0.00015289640103269625,
2116
+ "loss": 0.656,
2117
+ "step": 330
2118
+ },
2119
+ {
2120
+ "epoch": 0.99,
2121
+ "learning_rate": 0.00015262737668895405,
2122
+ "loss": 0.6649,
2123
+ "step": 331
2124
+ },
2125
+ {
2126
+ "epoch": 0.99,
2127
+ "learning_rate": 0.00015235782452288068,
2128
+ "loss": 0.6775,
2129
+ "step": 332
2130
+ },
2131
+ {
2132
+ "epoch": 0.99,
2133
+ "learning_rate": 0.00015208774723792943,
2134
+ "loss": 0.6637,
2135
+ "step": 333
2136
+ },
2137
+ {
2138
+ "epoch": 1.0,
2139
+ "learning_rate": 0.0001518171475428202,
2140
+ "loss": 0.7175,
2141
+ "step": 334
2142
+ }
2143
+ ],
2144
+ "logging_steps": 1,
2145
+ "max_steps": 1002,
2146
+ "num_train_epochs": 3,
2147
+ "save_steps": 500,
2148
+ "total_flos": 4.735056607476449e+17,
2149
+ "trial_name": null,
2150
+ "trial_params": null
2151
+ }
checkpoint-334/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7780122e038bb83236a6b213b013867a89c4e742cce8c1208c1a41fc54e6c57
3
+ size 4475
checkpoint-669/README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: True
10
+ - load_in_4bit: False
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.6.0.dev0
checkpoint-669/adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 16,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 32,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj",
20
+ "q_proj",
21
+ "down_proj",
22
+ "o_proj",
23
+ "up_proj",
24
+ "gate_proj",
25
+ "k_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
checkpoint-669/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:271457be4fbe175e6bc47cce64f06e88bdcd5e1c992e72f4162e5a5331961855
3
+ size 335705741
checkpoint-669/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18a2b1ed54334d6f9eab456f14f6162952de9ee6dc73054f8e9ab440749c7364
3
+ size 168624735
checkpoint-669/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1ab09093aa584365cebc0ff564820bbc3a15434c299d4b56d1d3b7eb8ec8d5b
3
+ size 14575
checkpoint-669/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a29e66a13e55b23150a2838301e660c7f48ed4ad4ac666ab03ba2e6758b9600e
3
+ size 627
checkpoint-669/trainer_state.json ADDED
@@ -0,0 +1,4297 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.998506348020911,
5
+ "eval_steps": 20,
6
+ "global_step": 669,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 2e-05,
14
+ "loss": 1.5197,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4e-05,
20
+ "loss": 1.5435,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 6e-05,
26
+ "loss": 1.3967,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 8e-05,
32
+ "loss": 1.368,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.01,
37
+ "learning_rate": 0.0001,
38
+ "loss": 1.2766,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 0.00012,
44
+ "loss": 1.1676,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.02,
49
+ "learning_rate": 0.00014,
50
+ "loss": 1.0507,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.02,
55
+ "learning_rate": 0.00016,
56
+ "loss": 1.0016,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.03,
61
+ "learning_rate": 0.00018,
62
+ "loss": 0.9427,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.03,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.9294,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.03,
73
+ "learning_rate": 0.00019999949852874652,
74
+ "loss": 0.9023,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.04,
79
+ "learning_rate": 0.00019999799412001546,
80
+ "loss": 0.8257,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.04,
85
+ "learning_rate": 0.0001999954867888952,
86
+ "loss": 0.9064,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.04,
91
+ "learning_rate": 0.00019999197656053288,
92
+ "loss": 0.8408,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.04,
97
+ "learning_rate": 0.00019998746347013398,
98
+ "loss": 0.9058,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.05,
103
+ "learning_rate": 0.0001999819475629623,
104
+ "loss": 0.8419,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.05,
109
+ "learning_rate": 0.00019997542889433915,
110
+ "loss": 0.8535,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.05,
115
+ "learning_rate": 0.00019996790752964305,
116
+ "loss": 0.8121,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.06,
121
+ "learning_rate": 0.00019995938354430898,
122
+ "loss": 0.7725,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.06,
127
+ "learning_rate": 0.00019994985702382758,
128
+ "loss": 0.7661,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.06,
133
+ "eval_loss": 1.517298698425293,
134
+ "eval_runtime": 15.7157,
135
+ "eval_samples_per_second": 37.097,
136
+ "eval_steps_per_second": 18.58,
137
+ "step": 20
138
+ },
139
+ {
140
+ "epoch": 0.06,
141
+ "learning_rate": 0.0001999393280637444,
142
+ "loss": 0.7878,
143
+ "step": 21
144
+ },
145
+ {
146
+ "epoch": 0.07,
147
+ "learning_rate": 0.00019992779676965885,
148
+ "loss": 0.7435,
149
+ "step": 22
150
+ },
151
+ {
152
+ "epoch": 0.07,
153
+ "learning_rate": 0.00019991526325722315,
154
+ "loss": 0.8219,
155
+ "step": 23
156
+ },
157
+ {
158
+ "epoch": 0.07,
159
+ "learning_rate": 0.00019990172765214128,
160
+ "loss": 0.7624,
161
+ "step": 24
162
+ },
163
+ {
164
+ "epoch": 0.07,
165
+ "learning_rate": 0.00019988719009016753,
166
+ "loss": 0.7527,
167
+ "step": 25
168
+ },
169
+ {
170
+ "epoch": 0.08,
171
+ "learning_rate": 0.00019987165071710527,
172
+ "loss": 0.7699,
173
+ "step": 26
174
+ },
175
+ {
176
+ "epoch": 0.08,
177
+ "learning_rate": 0.00019985510968880557,
178
+ "loss": 0.7536,
179
+ "step": 27
180
+ },
181
+ {
182
+ "epoch": 0.08,
183
+ "learning_rate": 0.00019983756717116536,
184
+ "loss": 0.7863,
185
+ "step": 28
186
+ },
187
+ {
188
+ "epoch": 0.09,
189
+ "learning_rate": 0.000199819023340126,
190
+ "loss": 0.7166,
191
+ "step": 29
192
+ },
193
+ {
194
+ "epoch": 0.09,
195
+ "learning_rate": 0.0001997994783816715,
196
+ "loss": 0.7459,
197
+ "step": 30
198
+ },
199
+ {
200
+ "epoch": 0.09,
201
+ "learning_rate": 0.00019977893249182652,
202
+ "loss": 0.7376,
203
+ "step": 31
204
+ },
205
+ {
206
+ "epoch": 0.1,
207
+ "learning_rate": 0.00019975738587665456,
208
+ "loss": 0.7624,
209
+ "step": 32
210
+ },
211
+ {
212
+ "epoch": 0.1,
213
+ "learning_rate": 0.00019973483875225574,
214
+ "loss": 0.7329,
215
+ "step": 33
216
+ },
217
+ {
218
+ "epoch": 0.1,
219
+ "learning_rate": 0.00019971129134476473,
220
+ "loss": 0.7512,
221
+ "step": 34
222
+ },
223
+ {
224
+ "epoch": 0.1,
225
+ "learning_rate": 0.0001996867438903486,
226
+ "loss": 0.7211,
227
+ "step": 35
228
+ },
229
+ {
230
+ "epoch": 0.11,
231
+ "learning_rate": 0.00019966119663520412,
232
+ "loss": 0.7467,
233
+ "step": 36
234
+ },
235
+ {
236
+ "epoch": 0.11,
237
+ "learning_rate": 0.00019963464983555556,
238
+ "loss": 0.7186,
239
+ "step": 37
240
+ },
241
+ {
242
+ "epoch": 0.11,
243
+ "learning_rate": 0.0001996071037576521,
244
+ "loss": 0.7244,
245
+ "step": 38
246
+ },
247
+ {
248
+ "epoch": 0.12,
249
+ "learning_rate": 0.00019957855867776508,
250
+ "loss": 0.7071,
251
+ "step": 39
252
+ },
253
+ {
254
+ "epoch": 0.12,
255
+ "learning_rate": 0.00019954901488218515,
256
+ "loss": 0.7681,
257
+ "step": 40
258
+ },
259
+ {
260
+ "epoch": 0.12,
261
+ "eval_loss": 1.2323185205459595,
262
+ "eval_runtime": 15.7174,
263
+ "eval_samples_per_second": 37.093,
264
+ "eval_steps_per_second": 18.578,
265
+ "step": 40
266
+ },
267
+ {
268
+ "epoch": 0.12,
269
+ "learning_rate": 0.0001995184726672197,
270
+ "loss": 0.708,
271
+ "step": 41
272
+ },
273
+ {
274
+ "epoch": 0.13,
275
+ "learning_rate": 0.00019948693233918952,
276
+ "loss": 0.7443,
277
+ "step": 42
278
+ },
279
+ {
280
+ "epoch": 0.13,
281
+ "learning_rate": 0.000199454394214426,
282
+ "loss": 0.7079,
283
+ "step": 43
284
+ },
285
+ {
286
+ "epoch": 0.13,
287
+ "learning_rate": 0.0001994208586192678,
288
+ "loss": 0.6812,
289
+ "step": 44
290
+ },
291
+ {
292
+ "epoch": 0.13,
293
+ "learning_rate": 0.0001993863258900577,
294
+ "loss": 0.6792,
295
+ "step": 45
296
+ },
297
+ {
298
+ "epoch": 0.14,
299
+ "learning_rate": 0.00019935079637313906,
300
+ "loss": 0.6553,
301
+ "step": 46
302
+ },
303
+ {
304
+ "epoch": 0.14,
305
+ "learning_rate": 0.0001993142704248525,
306
+ "loss": 0.6936,
307
+ "step": 47
308
+ },
309
+ {
310
+ "epoch": 0.14,
311
+ "learning_rate": 0.00019927674841153237,
312
+ "loss": 0.74,
313
+ "step": 48
314
+ },
315
+ {
316
+ "epoch": 0.15,
317
+ "learning_rate": 0.00019923823070950282,
318
+ "loss": 0.7154,
319
+ "step": 49
320
+ },
321
+ {
322
+ "epoch": 0.15,
323
+ "learning_rate": 0.0001991987177050743,
324
+ "loss": 0.6902,
325
+ "step": 50
326
+ },
327
+ {
328
+ "epoch": 0.15,
329
+ "learning_rate": 0.0001991582097945395,
330
+ "loss": 0.7009,
331
+ "step": 51
332
+ },
333
+ {
334
+ "epoch": 0.16,
335
+ "learning_rate": 0.00019911670738416947,
336
+ "loss": 0.699,
337
+ "step": 52
338
+ },
339
+ {
340
+ "epoch": 0.16,
341
+ "learning_rate": 0.00019907421089020955,
342
+ "loss": 0.681,
343
+ "step": 53
344
+ },
345
+ {
346
+ "epoch": 0.16,
347
+ "learning_rate": 0.00019903072073887507,
348
+ "loss": 0.7097,
349
+ "step": 54
350
+ },
351
+ {
352
+ "epoch": 0.16,
353
+ "learning_rate": 0.00019898623736634737,
354
+ "loss": 0.6861,
355
+ "step": 55
356
+ },
357
+ {
358
+ "epoch": 0.17,
359
+ "learning_rate": 0.000198940761218769,
360
+ "loss": 0.7279,
361
+ "step": 56
362
+ },
363
+ {
364
+ "epoch": 0.17,
365
+ "learning_rate": 0.00019889429275223957,
366
+ "loss": 0.6784,
367
+ "step": 57
368
+ },
369
+ {
370
+ "epoch": 0.17,
371
+ "learning_rate": 0.00019884683243281116,
372
+ "loss": 0.7,
373
+ "step": 58
374
+ },
375
+ {
376
+ "epoch": 0.18,
377
+ "learning_rate": 0.00019879838073648341,
378
+ "loss": 0.6634,
379
+ "step": 59
380
+ },
381
+ {
382
+ "epoch": 0.18,
383
+ "learning_rate": 0.00019874893814919906,
384
+ "loss": 0.6647,
385
+ "step": 60
386
+ },
387
+ {
388
+ "epoch": 0.18,
389
+ "eval_loss": 1.1306297779083252,
390
+ "eval_runtime": 15.73,
391
+ "eval_samples_per_second": 37.063,
392
+ "eval_steps_per_second": 18.563,
393
+ "step": 60
394
+ },
395
+ {
396
+ "epoch": 0.18,
397
+ "learning_rate": 0.00019869850516683875,
398
+ "loss": 0.7377,
399
+ "step": 61
400
+ },
401
+ {
402
+ "epoch": 0.19,
403
+ "learning_rate": 0.00019864708229521636,
404
+ "loss": 0.6794,
405
+ "step": 62
406
+ },
407
+ {
408
+ "epoch": 0.19,
409
+ "learning_rate": 0.00019859467005007368,
410
+ "loss": 0.6744,
411
+ "step": 63
412
+ },
413
+ {
414
+ "epoch": 0.19,
415
+ "learning_rate": 0.0001985412689570754,
416
+ "loss": 0.6921,
417
+ "step": 64
418
+ },
419
+ {
420
+ "epoch": 0.19,
421
+ "learning_rate": 0.00019848687955180386,
422
+ "loss": 0.7389,
423
+ "step": 65
424
+ },
425
+ {
426
+ "epoch": 0.2,
427
+ "learning_rate": 0.00019843150237975344,
428
+ "loss": 0.6795,
429
+ "step": 66
430
+ },
431
+ {
432
+ "epoch": 0.2,
433
+ "learning_rate": 0.00019837513799632534,
434
+ "loss": 0.6917,
435
+ "step": 67
436
+ },
437
+ {
438
+ "epoch": 0.2,
439
+ "learning_rate": 0.00019831778696682194,
440
+ "loss": 0.66,
441
+ "step": 68
442
+ },
443
+ {
444
+ "epoch": 0.21,
445
+ "learning_rate": 0.00019825944986644114,
446
+ "loss": 0.6926,
447
+ "step": 69
448
+ },
449
+ {
450
+ "epoch": 0.21,
451
+ "learning_rate": 0.00019820012728027044,
452
+ "loss": 0.7265,
453
+ "step": 70
454
+ },
455
+ {
456
+ "epoch": 0.21,
457
+ "learning_rate": 0.00019813981980328129,
458
+ "loss": 0.733,
459
+ "step": 71
460
+ },
461
+ {
462
+ "epoch": 0.22,
463
+ "learning_rate": 0.00019807852804032305,
464
+ "loss": 0.6543,
465
+ "step": 72
466
+ },
467
+ {
468
+ "epoch": 0.22,
469
+ "learning_rate": 0.00019801625260611685,
470
+ "loss": 0.673,
471
+ "step": 73
472
+ },
473
+ {
474
+ "epoch": 0.22,
475
+ "learning_rate": 0.00019795299412524945,
476
+ "loss": 0.6773,
477
+ "step": 74
478
+ },
479
+ {
480
+ "epoch": 0.22,
481
+ "learning_rate": 0.0001978887532321671,
482
+ "loss": 0.6469,
483
+ "step": 75
484
+ },
485
+ {
486
+ "epoch": 0.23,
487
+ "learning_rate": 0.000197823530571169,
488
+ "loss": 0.7272,
489
+ "step": 76
490
+ },
491
+ {
492
+ "epoch": 0.23,
493
+ "learning_rate": 0.00019775732679640093,
494
+ "loss": 0.7015,
495
+ "step": 77
496
+ },
497
+ {
498
+ "epoch": 0.23,
499
+ "learning_rate": 0.0001976901425718487,
500
+ "loss": 0.6973,
501
+ "step": 78
502
+ },
503
+ {
504
+ "epoch": 0.24,
505
+ "learning_rate": 0.00019762197857133148,
506
+ "loss": 0.6831,
507
+ "step": 79
508
+ },
509
+ {
510
+ "epoch": 0.24,
511
+ "learning_rate": 0.00019755283547849494,
512
+ "loss": 0.6742,
513
+ "step": 80
514
+ },
515
+ {
516
+ "epoch": 0.24,
517
+ "eval_loss": 1.0846511125564575,
518
+ "eval_runtime": 15.6885,
519
+ "eval_samples_per_second": 37.161,
520
+ "eval_steps_per_second": 18.612,
521
+ "step": 80
522
+ },
523
+ {
524
+ "epoch": 0.24,
525
+ "learning_rate": 0.0001974827139868046,
526
+ "loss": 0.683,
527
+ "step": 81
528
+ },
529
+ {
530
+ "epoch": 0.24,
531
+ "learning_rate": 0.0001974116147995387,
532
+ "loss": 0.7063,
533
+ "step": 82
534
+ },
535
+ {
536
+ "epoch": 0.25,
537
+ "learning_rate": 0.00019733953862978123,
538
+ "loss": 0.6963,
539
+ "step": 83
540
+ },
541
+ {
542
+ "epoch": 0.25,
543
+ "learning_rate": 0.00019726648620041468,
544
+ "loss": 0.7179,
545
+ "step": 84
546
+ },
547
+ {
548
+ "epoch": 0.25,
549
+ "learning_rate": 0.00019719245824411296,
550
+ "loss": 0.7315,
551
+ "step": 85
552
+ },
553
+ {
554
+ "epoch": 0.26,
555
+ "learning_rate": 0.0001971174555033339,
556
+ "loss": 0.6839,
557
+ "step": 86
558
+ },
559
+ {
560
+ "epoch": 0.26,
561
+ "learning_rate": 0.00019704147873031186,
562
+ "loss": 0.6917,
563
+ "step": 87
564
+ },
565
+ {
566
+ "epoch": 0.26,
567
+ "learning_rate": 0.00019696452868705024,
568
+ "loss": 0.6828,
569
+ "step": 88
570
+ },
571
+ {
572
+ "epoch": 0.27,
573
+ "learning_rate": 0.00019688660614531367,
574
+ "loss": 0.6965,
575
+ "step": 89
576
+ },
577
+ {
578
+ "epoch": 0.27,
579
+ "learning_rate": 0.00019680771188662044,
580
+ "loss": 0.6522,
581
+ "step": 90
582
+ },
583
+ {
584
+ "epoch": 0.27,
585
+ "learning_rate": 0.00019672784670223464,
586
+ "loss": 0.6939,
587
+ "step": 91
588
+ },
589
+ {
590
+ "epoch": 0.27,
591
+ "learning_rate": 0.0001966470113931582,
592
+ "loss": 0.7037,
593
+ "step": 92
594
+ },
595
+ {
596
+ "epoch": 0.28,
597
+ "learning_rate": 0.00019656520677012264,
598
+ "loss": 0.6903,
599
+ "step": 93
600
+ },
601
+ {
602
+ "epoch": 0.28,
603
+ "learning_rate": 0.00019648243365358146,
604
+ "loss": 0.706,
605
+ "step": 94
606
+ },
607
+ {
608
+ "epoch": 0.28,
609
+ "learning_rate": 0.0001963986928737014,
610
+ "loss": 0.6253,
611
+ "step": 95
612
+ },
613
+ {
614
+ "epoch": 0.29,
615
+ "learning_rate": 0.00019631398527035422,
616
+ "loss": 0.732,
617
+ "step": 96
618
+ },
619
+ {
620
+ "epoch": 0.29,
621
+ "learning_rate": 0.0001962283116931086,
622
+ "loss": 0.6935,
623
+ "step": 97
624
+ },
625
+ {
626
+ "epoch": 0.29,
627
+ "learning_rate": 0.00019614167300122126,
628
+ "loss": 0.731,
629
+ "step": 98
630
+ },
631
+ {
632
+ "epoch": 0.3,
633
+ "learning_rate": 0.0001960540700636284,
634
+ "loss": 0.6916,
635
+ "step": 99
636
+ },
637
+ {
638
+ "epoch": 0.3,
639
+ "learning_rate": 0.0001959655037589372,
640
+ "loss": 0.6995,
641
+ "step": 100
642
+ },
643
+ {
644
+ "epoch": 0.3,
645
+ "eval_loss": 1.0572510957717896,
646
+ "eval_runtime": 15.7086,
647
+ "eval_samples_per_second": 37.113,
648
+ "eval_steps_per_second": 18.588,
649
+ "step": 100
650
+ },
651
+ {
652
+ "epoch": 0.3,
653
+ "learning_rate": 0.0001958759749754167,
654
+ "loss": 0.7313,
655
+ "step": 101
656
+ },
657
+ {
658
+ "epoch": 0.3,
659
+ "learning_rate": 0.00019578548461098914,
660
+ "loss": 0.7337,
661
+ "step": 102
662
+ },
663
+ {
664
+ "epoch": 0.31,
665
+ "learning_rate": 0.0001956940335732209,
666
+ "loss": 0.6721,
667
+ "step": 103
668
+ },
669
+ {
670
+ "epoch": 0.31,
671
+ "learning_rate": 0.00019560162277931325,
672
+ "loss": 0.6907,
673
+ "step": 104
674
+ },
675
+ {
676
+ "epoch": 0.31,
677
+ "learning_rate": 0.0001955082531560934,
678
+ "loss": 0.6599,
679
+ "step": 105
680
+ },
681
+ {
682
+ "epoch": 0.32,
683
+ "learning_rate": 0.00019541392564000488,
684
+ "loss": 0.6941,
685
+ "step": 106
686
+ },
687
+ {
688
+ "epoch": 0.32,
689
+ "learning_rate": 0.00019531864117709855,
690
+ "loss": 0.7037,
691
+ "step": 107
692
+ },
693
+ {
694
+ "epoch": 0.32,
695
+ "learning_rate": 0.00019522240072302274,
696
+ "loss": 0.6819,
697
+ "step": 108
698
+ },
699
+ {
700
+ "epoch": 0.33,
701
+ "learning_rate": 0.00019512520524301387,
702
+ "loss": 0.6729,
703
+ "step": 109
704
+ },
705
+ {
706
+ "epoch": 0.33,
707
+ "learning_rate": 0.00019502705571188672,
708
+ "loss": 0.713,
709
+ "step": 110
710
+ },
711
+ {
712
+ "epoch": 0.33,
713
+ "learning_rate": 0.00019492795311402473,
714
+ "loss": 0.6638,
715
+ "step": 111
716
+ },
717
+ {
718
+ "epoch": 0.33,
719
+ "learning_rate": 0.0001948278984433699,
720
+ "loss": 0.647,
721
+ "step": 112
722
+ },
723
+ {
724
+ "epoch": 0.34,
725
+ "learning_rate": 0.00019472689270341303,
726
+ "loss": 0.614,
727
+ "step": 113
728
+ },
729
+ {
730
+ "epoch": 0.34,
731
+ "learning_rate": 0.0001946249369071837,
732
+ "loss": 0.6783,
733
+ "step": 114
734
+ },
735
+ {
736
+ "epoch": 0.34,
737
+ "learning_rate": 0.0001945220320772399,
738
+ "loss": 0.7093,
739
+ "step": 115
740
+ },
741
+ {
742
+ "epoch": 0.35,
743
+ "learning_rate": 0.00019441817924565786,
744
+ "loss": 0.7179,
745
+ "step": 116
746
+ },
747
+ {
748
+ "epoch": 0.35,
749
+ "learning_rate": 0.00019431337945402185,
750
+ "loss": 0.7241,
751
+ "step": 117
752
+ },
753
+ {
754
+ "epoch": 0.35,
755
+ "learning_rate": 0.0001942076337534135,
756
+ "loss": 0.7168,
757
+ "step": 118
758
+ },
759
+ {
760
+ "epoch": 0.36,
761
+ "learning_rate": 0.00019410094320440138,
762
+ "loss": 0.7127,
763
+ "step": 119
764
+ },
765
+ {
766
+ "epoch": 0.36,
767
+ "learning_rate": 0.00019399330887703037,
768
+ "loss": 0.6883,
769
+ "step": 120
770
+ },
771
+ {
772
+ "epoch": 0.36,
773
+ "eval_loss": 1.0412259101867676,
774
+ "eval_runtime": 15.6448,
775
+ "eval_samples_per_second": 37.265,
776
+ "eval_steps_per_second": 18.664,
777
+ "step": 120
778
+ },
779
+ {
780
+ "epoch": 0.36,
781
+ "learning_rate": 0.00019388473185081087,
782
+ "loss": 0.6663,
783
+ "step": 121
784
+ },
785
+ {
786
+ "epoch": 0.36,
787
+ "learning_rate": 0.00019377521321470805,
788
+ "loss": 0.7319,
789
+ "step": 122
790
+ },
791
+ {
792
+ "epoch": 0.37,
793
+ "learning_rate": 0.00019366475406713086,
794
+ "loss": 0.6942,
795
+ "step": 123
796
+ },
797
+ {
798
+ "epoch": 0.37,
799
+ "learning_rate": 0.00019355335551592105,
800
+ "loss": 0.6847,
801
+ "step": 124
802
+ },
803
+ {
804
+ "epoch": 0.37,
805
+ "learning_rate": 0.000193441018678342,
806
+ "loss": 0.6732,
807
+ "step": 125
808
+ },
809
+ {
810
+ "epoch": 0.38,
811
+ "learning_rate": 0.00019332774468106768,
812
+ "loss": 0.6661,
813
+ "step": 126
814
+ },
815
+ {
816
+ "epoch": 0.38,
817
+ "learning_rate": 0.00019321353466017108,
818
+ "loss": 0.7044,
819
+ "step": 127
820
+ },
821
+ {
822
+ "epoch": 0.38,
823
+ "learning_rate": 0.00019309838976111311,
824
+ "loss": 0.6794,
825
+ "step": 128
826
+ },
827
+ {
828
+ "epoch": 0.39,
829
+ "learning_rate": 0.00019298231113873087,
830
+ "loss": 0.7275,
831
+ "step": 129
832
+ },
833
+ {
834
+ "epoch": 0.39,
835
+ "learning_rate": 0.00019286529995722623,
836
+ "loss": 0.685,
837
+ "step": 130
838
+ },
839
+ {
840
+ "epoch": 0.39,
841
+ "learning_rate": 0.00019274735739015404,
842
+ "loss": 0.6825,
843
+ "step": 131
844
+ },
845
+ {
846
+ "epoch": 0.39,
847
+ "learning_rate": 0.00019262848462041045,
848
+ "loss": 0.6857,
849
+ "step": 132
850
+ },
851
+ {
852
+ "epoch": 0.4,
853
+ "learning_rate": 0.000192508682840221,
854
+ "loss": 0.7139,
855
+ "step": 133
856
+ },
857
+ {
858
+ "epoch": 0.4,
859
+ "learning_rate": 0.0001923879532511287,
860
+ "loss": 0.6413,
861
+ "step": 134
862
+ },
863
+ {
864
+ "epoch": 0.4,
865
+ "learning_rate": 0.00019226629706398186,
866
+ "loss": 0.7257,
867
+ "step": 135
868
+ },
869
+ {
870
+ "epoch": 0.41,
871
+ "learning_rate": 0.0001921437154989221,
872
+ "loss": 0.6365,
873
+ "step": 136
874
+ },
875
+ {
876
+ "epoch": 0.41,
877
+ "learning_rate": 0.00019202020978537207,
878
+ "loss": 0.66,
879
+ "step": 137
880
+ },
881
+ {
882
+ "epoch": 0.41,
883
+ "learning_rate": 0.00019189578116202307,
884
+ "loss": 0.6525,
885
+ "step": 138
886
+ },
887
+ {
888
+ "epoch": 0.42,
889
+ "learning_rate": 0.00019177043087682265,
890
+ "loss": 0.6438,
891
+ "step": 139
892
+ },
893
+ {
894
+ "epoch": 0.42,
895
+ "learning_rate": 0.00019164416018696207,
896
+ "loss": 0.6437,
897
+ "step": 140
898
+ },
899
+ {
900
+ "epoch": 0.42,
901
+ "eval_loss": 1.0374538898468018,
902
+ "eval_runtime": 15.6903,
903
+ "eval_samples_per_second": 37.157,
904
+ "eval_steps_per_second": 18.61,
905
+ "step": 140
906
+ },
907
+ {
908
+ "epoch": 0.42,
909
+ "learning_rate": 0.00019151697035886382,
910
+ "loss": 0.6878,
911
+ "step": 141
912
+ },
913
+ {
914
+ "epoch": 0.42,
915
+ "learning_rate": 0.00019138886266816866,
916
+ "loss": 0.7072,
917
+ "step": 142
918
+ },
919
+ {
920
+ "epoch": 0.43,
921
+ "learning_rate": 0.00019125983839972317,
922
+ "loss": 0.69,
923
+ "step": 143
924
+ },
925
+ {
926
+ "epoch": 0.43,
927
+ "learning_rate": 0.00019112989884756653,
928
+ "loss": 0.7666,
929
+ "step": 144
930
+ },
931
+ {
932
+ "epoch": 0.43,
933
+ "learning_rate": 0.00019099904531491774,
934
+ "loss": 0.689,
935
+ "step": 145
936
+ },
937
+ {
938
+ "epoch": 0.44,
939
+ "learning_rate": 0.0001908672791141625,
940
+ "loss": 0.7394,
941
+ "step": 146
942
+ },
943
+ {
944
+ "epoch": 0.44,
945
+ "learning_rate": 0.0001907346015668401,
946
+ "loss": 0.6882,
947
+ "step": 147
948
+ },
949
+ {
950
+ "epoch": 0.44,
951
+ "learning_rate": 0.00019060101400362998,
952
+ "loss": 0.6394,
953
+ "step": 148
954
+ },
955
+ {
956
+ "epoch": 0.45,
957
+ "learning_rate": 0.00019046651776433862,
958
+ "loss": 0.6932,
959
+ "step": 149
960
+ },
961
+ {
962
+ "epoch": 0.45,
963
+ "learning_rate": 0.00019033111419788597,
964
+ "loss": 0.6676,
965
+ "step": 150
966
+ },
967
+ {
968
+ "epoch": 0.45,
969
+ "learning_rate": 0.00019019480466229201,
970
+ "loss": 0.6818,
971
+ "step": 151
972
+ },
973
+ {
974
+ "epoch": 0.45,
975
+ "learning_rate": 0.000190057590524663,
976
+ "loss": 0.6695,
977
+ "step": 152
978
+ },
979
+ {
980
+ "epoch": 0.46,
981
+ "learning_rate": 0.0001899194731611778,
982
+ "loss": 0.7082,
983
+ "step": 153
984
+ },
985
+ {
986
+ "epoch": 0.46,
987
+ "learning_rate": 0.00018978045395707418,
988
+ "loss": 0.6626,
989
+ "step": 154
990
+ },
991
+ {
992
+ "epoch": 0.46,
993
+ "learning_rate": 0.00018964053430663483,
994
+ "loss": 0.6708,
995
+ "step": 155
996
+ },
997
+ {
998
+ "epoch": 0.47,
999
+ "learning_rate": 0.0001894997156131734,
1000
+ "loss": 0.7864,
1001
+ "step": 156
1002
+ },
1003
+ {
1004
+ "epoch": 0.47,
1005
+ "learning_rate": 0.00018935799928902045,
1006
+ "loss": 0.6899,
1007
+ "step": 157
1008
+ },
1009
+ {
1010
+ "epoch": 0.47,
1011
+ "learning_rate": 0.0001892153867555092,
1012
+ "loss": 0.6574,
1013
+ "step": 158
1014
+ },
1015
+ {
1016
+ "epoch": 0.47,
1017
+ "learning_rate": 0.00018907187944296137,
1018
+ "loss": 0.7157,
1019
+ "step": 159
1020
+ },
1021
+ {
1022
+ "epoch": 0.48,
1023
+ "learning_rate": 0.00018892747879067286,
1024
+ "loss": 0.6331,
1025
+ "step": 160
1026
+ },
1027
+ {
1028
+ "epoch": 0.48,
1029
+ "eval_loss": 1.0186299085617065,
1030
+ "eval_runtime": 15.6809,
1031
+ "eval_samples_per_second": 37.179,
1032
+ "eval_steps_per_second": 18.621,
1033
+ "step": 160
1034
+ },
1035
+ {
1036
+ "epoch": 0.48,
1037
+ "learning_rate": 0.0001887821862468991,
1038
+ "loss": 0.7219,
1039
+ "step": 161
1040
+ },
1041
+ {
1042
+ "epoch": 0.48,
1043
+ "learning_rate": 0.00018863600326884082,
1044
+ "loss": 0.7203,
1045
+ "step": 162
1046
+ },
1047
+ {
1048
+ "epoch": 0.49,
1049
+ "learning_rate": 0.00018848893132262926,
1050
+ "loss": 0.6722,
1051
+ "step": 163
1052
+ },
1053
+ {
1054
+ "epoch": 0.49,
1055
+ "learning_rate": 0.00018834097188331143,
1056
+ "loss": 0.7223,
1057
+ "step": 164
1058
+ },
1059
+ {
1060
+ "epoch": 0.49,
1061
+ "learning_rate": 0.0001881921264348355,
1062
+ "loss": 0.659,
1063
+ "step": 165
1064
+ },
1065
+ {
1066
+ "epoch": 0.5,
1067
+ "learning_rate": 0.00018804239647003573,
1068
+ "loss": 0.6763,
1069
+ "step": 166
1070
+ },
1071
+ {
1072
+ "epoch": 0.5,
1073
+ "learning_rate": 0.00018789178349061753,
1074
+ "loss": 0.649,
1075
+ "step": 167
1076
+ },
1077
+ {
1078
+ "epoch": 0.5,
1079
+ "learning_rate": 0.00018774028900714256,
1080
+ "loss": 0.6762,
1081
+ "step": 168
1082
+ },
1083
+ {
1084
+ "epoch": 0.5,
1085
+ "learning_rate": 0.00018758791453901336,
1086
+ "loss": 0.6547,
1087
+ "step": 169
1088
+ },
1089
+ {
1090
+ "epoch": 0.51,
1091
+ "learning_rate": 0.00018743466161445823,
1092
+ "loss": 0.6516,
1093
+ "step": 170
1094
+ },
1095
+ {
1096
+ "epoch": 0.51,
1097
+ "learning_rate": 0.0001872805317705159,
1098
+ "loss": 0.6479,
1099
+ "step": 171
1100
+ },
1101
+ {
1102
+ "epoch": 0.51,
1103
+ "learning_rate": 0.0001871255265530201,
1104
+ "loss": 0.6986,
1105
+ "step": 172
1106
+ },
1107
+ {
1108
+ "epoch": 0.52,
1109
+ "learning_rate": 0.00018696964751658404,
1110
+ "loss": 0.7402,
1111
+ "step": 173
1112
+ },
1113
+ {
1114
+ "epoch": 0.52,
1115
+ "learning_rate": 0.00018681289622458485,
1116
+ "loss": 0.6629,
1117
+ "step": 174
1118
+ },
1119
+ {
1120
+ "epoch": 0.52,
1121
+ "learning_rate": 0.00018665527424914787,
1122
+ "loss": 0.6761,
1123
+ "step": 175
1124
+ },
1125
+ {
1126
+ "epoch": 0.53,
1127
+ "learning_rate": 0.00018649678317113084,
1128
+ "loss": 0.6746,
1129
+ "step": 176
1130
+ },
1131
+ {
1132
+ "epoch": 0.53,
1133
+ "learning_rate": 0.0001863374245801082,
1134
+ "loss": 0.67,
1135
+ "step": 177
1136
+ },
1137
+ {
1138
+ "epoch": 0.53,
1139
+ "learning_rate": 0.00018617720007435497,
1140
+ "loss": 0.7406,
1141
+ "step": 178
1142
+ },
1143
+ {
1144
+ "epoch": 0.53,
1145
+ "learning_rate": 0.00018601611126083087,
1146
+ "loss": 0.6735,
1147
+ "step": 179
1148
+ },
1149
+ {
1150
+ "epoch": 0.54,
1151
+ "learning_rate": 0.000185854159755164,
1152
+ "loss": 0.6686,
1153
+ "step": 180
1154
+ },
1155
+ {
1156
+ "epoch": 0.54,
1157
+ "eval_loss": 1.0153461694717407,
1158
+ "eval_runtime": 15.6769,
1159
+ "eval_samples_per_second": 37.189,
1160
+ "eval_steps_per_second": 18.626,
1161
+ "step": 180
1162
+ },
1163
+ {
1164
+ "epoch": 0.54,
1165
+ "learning_rate": 0.00018569134718163496,
1166
+ "loss": 0.6416,
1167
+ "step": 181
1168
+ },
1169
+ {
1170
+ "epoch": 0.54,
1171
+ "learning_rate": 0.00018552767517316022,
1172
+ "loss": 0.6456,
1173
+ "step": 182
1174
+ },
1175
+ {
1176
+ "epoch": 0.55,
1177
+ "learning_rate": 0.00018536314537127586,
1178
+ "loss": 0.6794,
1179
+ "step": 183
1180
+ },
1181
+ {
1182
+ "epoch": 0.55,
1183
+ "learning_rate": 0.00018519775942612128,
1184
+ "loss": 0.6754,
1185
+ "step": 184
1186
+ },
1187
+ {
1188
+ "epoch": 0.55,
1189
+ "learning_rate": 0.00018503151899642236,
1190
+ "loss": 0.6251,
1191
+ "step": 185
1192
+ },
1193
+ {
1194
+ "epoch": 0.56,
1195
+ "learning_rate": 0.00018486442574947511,
1196
+ "loss": 0.7354,
1197
+ "step": 186
1198
+ },
1199
+ {
1200
+ "epoch": 0.56,
1201
+ "learning_rate": 0.00018469648136112867,
1202
+ "loss": 0.6457,
1203
+ "step": 187
1204
+ },
1205
+ {
1206
+ "epoch": 0.56,
1207
+ "learning_rate": 0.0001845276875157687,
1208
+ "loss": 0.6823,
1209
+ "step": 188
1210
+ },
1211
+ {
1212
+ "epoch": 0.56,
1213
+ "learning_rate": 0.00018435804590630043,
1214
+ "loss": 0.6891,
1215
+ "step": 189
1216
+ },
1217
+ {
1218
+ "epoch": 0.57,
1219
+ "learning_rate": 0.0001841875582341317,
1220
+ "loss": 0.6731,
1221
+ "step": 190
1222
+ },
1223
+ {
1224
+ "epoch": 0.57,
1225
+ "learning_rate": 0.00018401622620915583,
1226
+ "loss": 0.6857,
1227
+ "step": 191
1228
+ },
1229
+ {
1230
+ "epoch": 0.57,
1231
+ "learning_rate": 0.0001838440515497345,
1232
+ "loss": 0.6807,
1233
+ "step": 192
1234
+ },
1235
+ {
1236
+ "epoch": 0.58,
1237
+ "learning_rate": 0.0001836710359826806,
1238
+ "loss": 0.6622,
1239
+ "step": 193
1240
+ },
1241
+ {
1242
+ "epoch": 0.58,
1243
+ "learning_rate": 0.00018349718124324076,
1244
+ "loss": 0.6628,
1245
+ "step": 194
1246
+ },
1247
+ {
1248
+ "epoch": 0.58,
1249
+ "learning_rate": 0.00018332248907507807,
1250
+ "loss": 0.724,
1251
+ "step": 195
1252
+ },
1253
+ {
1254
+ "epoch": 0.59,
1255
+ "learning_rate": 0.00018314696123025454,
1256
+ "loss": 0.675,
1257
+ "step": 196
1258
+ },
1259
+ {
1260
+ "epoch": 0.59,
1261
+ "learning_rate": 0.00018297059946921355,
1262
+ "loss": 0.677,
1263
+ "step": 197
1264
+ },
1265
+ {
1266
+ "epoch": 0.59,
1267
+ "learning_rate": 0.00018279340556076216,
1268
+ "loss": 0.6555,
1269
+ "step": 198
1270
+ },
1271
+ {
1272
+ "epoch": 0.59,
1273
+ "learning_rate": 0.0001826153812820534,
1274
+ "loss": 0.6273,
1275
+ "step": 199
1276
+ },
1277
+ {
1278
+ "epoch": 0.6,
1279
+ "learning_rate": 0.0001824365284185684,
1280
+ "loss": 0.6767,
1281
+ "step": 200
1282
+ },
1283
+ {
1284
+ "epoch": 0.6,
1285
+ "eval_loss": 1.004173994064331,
1286
+ "eval_runtime": 15.726,
1287
+ "eval_samples_per_second": 37.072,
1288
+ "eval_steps_per_second": 18.568,
1289
+ "step": 200
1290
+ },
1291
+ {
1292
+ "epoch": 0.6,
1293
+ "learning_rate": 0.00018225684876409863,
1294
+ "loss": 0.6719,
1295
+ "step": 201
1296
+ },
1297
+ {
1298
+ "epoch": 0.6,
1299
+ "learning_rate": 0.00018207634412072764,
1300
+ "loss": 0.6522,
1301
+ "step": 202
1302
+ },
1303
+ {
1304
+ "epoch": 0.61,
1305
+ "learning_rate": 0.0001818950162988133,
1306
+ "loss": 0.6957,
1307
+ "step": 203
1308
+ },
1309
+ {
1310
+ "epoch": 0.61,
1311
+ "learning_rate": 0.00018171286711696934,
1312
+ "loss": 0.7013,
1313
+ "step": 204
1314
+ },
1315
+ {
1316
+ "epoch": 0.61,
1317
+ "learning_rate": 0.00018152989840204735,
1318
+ "loss": 0.6603,
1319
+ "step": 205
1320
+ },
1321
+ {
1322
+ "epoch": 0.62,
1323
+ "learning_rate": 0.0001813461119891184,
1324
+ "loss": 0.6788,
1325
+ "step": 206
1326
+ },
1327
+ {
1328
+ "epoch": 0.62,
1329
+ "learning_rate": 0.0001811615097214545,
1330
+ "loss": 0.6795,
1331
+ "step": 207
1332
+ },
1333
+ {
1334
+ "epoch": 0.62,
1335
+ "learning_rate": 0.00018097609345051025,
1336
+ "loss": 0.6458,
1337
+ "step": 208
1338
+ },
1339
+ {
1340
+ "epoch": 0.62,
1341
+ "learning_rate": 0.00018078986503590428,
1342
+ "loss": 0.6741,
1343
+ "step": 209
1344
+ },
1345
+ {
1346
+ "epoch": 0.63,
1347
+ "learning_rate": 0.00018060282634540053,
1348
+ "loss": 0.6537,
1349
+ "step": 210
1350
+ },
1351
+ {
1352
+ "epoch": 0.63,
1353
+ "learning_rate": 0.0001804149792548895,
1354
+ "loss": 0.6501,
1355
+ "step": 211
1356
+ },
1357
+ {
1358
+ "epoch": 0.63,
1359
+ "learning_rate": 0.00018022632564836948,
1360
+ "loss": 0.673,
1361
+ "step": 212
1362
+ },
1363
+ {
1364
+ "epoch": 0.64,
1365
+ "learning_rate": 0.00018003686741792772,
1366
+ "loss": 0.6746,
1367
+ "step": 213
1368
+ },
1369
+ {
1370
+ "epoch": 0.64,
1371
+ "learning_rate": 0.0001798466064637214,
1372
+ "loss": 0.6329,
1373
+ "step": 214
1374
+ },
1375
+ {
1376
+ "epoch": 0.64,
1377
+ "learning_rate": 0.00017965554469395837,
1378
+ "loss": 0.69,
1379
+ "step": 215
1380
+ },
1381
+ {
1382
+ "epoch": 0.65,
1383
+ "learning_rate": 0.00017946368402487845,
1384
+ "loss": 0.6513,
1385
+ "step": 216
1386
+ },
1387
+ {
1388
+ "epoch": 0.65,
1389
+ "learning_rate": 0.00017927102638073383,
1390
+ "loss": 0.6758,
1391
+ "step": 217
1392
+ },
1393
+ {
1394
+ "epoch": 0.65,
1395
+ "learning_rate": 0.00017907757369376985,
1396
+ "loss": 0.6776,
1397
+ "step": 218
1398
+ },
1399
+ {
1400
+ "epoch": 0.65,
1401
+ "learning_rate": 0.00017888332790420584,
1402
+ "loss": 0.6487,
1403
+ "step": 219
1404
+ },
1405
+ {
1406
+ "epoch": 0.66,
1407
+ "learning_rate": 0.00017868829096021527,
1408
+ "loss": 0.7037,
1409
+ "step": 220
1410
+ },
1411
+ {
1412
+ "epoch": 0.66,
1413
+ "eval_loss": 1.002336025238037,
1414
+ "eval_runtime": 17.4124,
1415
+ "eval_samples_per_second": 33.482,
1416
+ "eval_steps_per_second": 16.77,
1417
+ "step": 220
1418
+ },
1419
+ {
1420
+ "epoch": 0.66,
1421
+ "learning_rate": 0.00017849246481790664,
1422
+ "loss": 0.71,
1423
+ "step": 221
1424
+ },
1425
+ {
1426
+ "epoch": 0.66,
1427
+ "learning_rate": 0.00017829585144130356,
1428
+ "loss": 0.6695,
1429
+ "step": 222
1430
+ },
1431
+ {
1432
+ "epoch": 0.67,
1433
+ "learning_rate": 0.00017809845280232514,
1434
+ "loss": 0.6645,
1435
+ "step": 223
1436
+ },
1437
+ {
1438
+ "epoch": 0.67,
1439
+ "learning_rate": 0.0001779002708807662,
1440
+ "loss": 0.6983,
1441
+ "step": 224
1442
+ },
1443
+ {
1444
+ "epoch": 0.67,
1445
+ "learning_rate": 0.0001777013076642776,
1446
+ "loss": 0.6785,
1447
+ "step": 225
1448
+ },
1449
+ {
1450
+ "epoch": 0.68,
1451
+ "learning_rate": 0.0001775015651483459,
1452
+ "loss": 0.6791,
1453
+ "step": 226
1454
+ },
1455
+ {
1456
+ "epoch": 0.68,
1457
+ "learning_rate": 0.0001773010453362737,
1458
+ "loss": 0.695,
1459
+ "step": 227
1460
+ },
1461
+ {
1462
+ "epoch": 0.68,
1463
+ "learning_rate": 0.00017709975023915949,
1464
+ "loss": 0.6352,
1465
+ "step": 228
1466
+ },
1467
+ {
1468
+ "epoch": 0.68,
1469
+ "learning_rate": 0.00017689768187587732,
1470
+ "loss": 0.6911,
1471
+ "step": 229
1472
+ },
1473
+ {
1474
+ "epoch": 0.69,
1475
+ "learning_rate": 0.0001766948422730567,
1476
+ "loss": 0.7189,
1477
+ "step": 230
1478
+ },
1479
+ {
1480
+ "epoch": 0.69,
1481
+ "learning_rate": 0.00017649123346506223,
1482
+ "loss": 0.6838,
1483
+ "step": 231
1484
+ },
1485
+ {
1486
+ "epoch": 0.69,
1487
+ "learning_rate": 0.0001762868574939732,
1488
+ "loss": 0.7209,
1489
+ "step": 232
1490
+ },
1491
+ {
1492
+ "epoch": 0.7,
1493
+ "learning_rate": 0.00017608171640956313,
1494
+ "loss": 0.6719,
1495
+ "step": 233
1496
+ },
1497
+ {
1498
+ "epoch": 0.7,
1499
+ "learning_rate": 0.0001758758122692791,
1500
+ "loss": 0.6624,
1501
+ "step": 234
1502
+ },
1503
+ {
1504
+ "epoch": 0.7,
1505
+ "learning_rate": 0.0001756691471382213,
1506
+ "loss": 0.637,
1507
+ "step": 235
1508
+ },
1509
+ {
1510
+ "epoch": 0.71,
1511
+ "learning_rate": 0.00017546172308912213,
1512
+ "loss": 0.653,
1513
+ "step": 236
1514
+ },
1515
+ {
1516
+ "epoch": 0.71,
1517
+ "learning_rate": 0.0001752535422023256,
1518
+ "loss": 0.6368,
1519
+ "step": 237
1520
+ },
1521
+ {
1522
+ "epoch": 0.71,
1523
+ "learning_rate": 0.00017504460656576627,
1524
+ "loss": 0.6713,
1525
+ "step": 238
1526
+ },
1527
+ {
1528
+ "epoch": 0.71,
1529
+ "learning_rate": 0.00017483491827494847,
1530
+ "loss": 0.6614,
1531
+ "step": 239
1532
+ },
1533
+ {
1534
+ "epoch": 0.72,
1535
+ "learning_rate": 0.0001746244794329252,
1536
+ "loss": 0.6994,
1537
+ "step": 240
1538
+ },
1539
+ {
1540
+ "epoch": 0.72,
1541
+ "eval_loss": 1.0014034509658813,
1542
+ "eval_runtime": 15.6575,
1543
+ "eval_samples_per_second": 37.235,
1544
+ "eval_steps_per_second": 18.649,
1545
+ "step": 240
1546
+ },
1547
+ {
1548
+ "epoch": 0.72,
1549
+ "learning_rate": 0.00017441329215027706,
1550
+ "loss": 0.6644,
1551
+ "step": 241
1552
+ },
1553
+ {
1554
+ "epoch": 0.72,
1555
+ "learning_rate": 0.0001742013585450911,
1556
+ "loss": 0.6374,
1557
+ "step": 242
1558
+ },
1559
+ {
1560
+ "epoch": 0.73,
1561
+ "learning_rate": 0.00017398868074293948,
1562
+ "loss": 0.6627,
1563
+ "step": 243
1564
+ },
1565
+ {
1566
+ "epoch": 0.73,
1567
+ "learning_rate": 0.00017377526087685832,
1568
+ "loss": 0.6936,
1569
+ "step": 244
1570
+ },
1571
+ {
1572
+ "epoch": 0.73,
1573
+ "learning_rate": 0.00017356110108732617,
1574
+ "loss": 0.6419,
1575
+ "step": 245
1576
+ },
1577
+ {
1578
+ "epoch": 0.73,
1579
+ "learning_rate": 0.0001733462035222426,
1580
+ "loss": 0.6402,
1581
+ "step": 246
1582
+ },
1583
+ {
1584
+ "epoch": 0.74,
1585
+ "learning_rate": 0.0001731305703369066,
1586
+ "loss": 0.7028,
1587
+ "step": 247
1588
+ },
1589
+ {
1590
+ "epoch": 0.74,
1591
+ "learning_rate": 0.0001729142036939951,
1592
+ "loss": 0.6146,
1593
+ "step": 248
1594
+ },
1595
+ {
1596
+ "epoch": 0.74,
1597
+ "learning_rate": 0.0001726971057635411,
1598
+ "loss": 0.6702,
1599
+ "step": 249
1600
+ },
1601
+ {
1602
+ "epoch": 0.75,
1603
+ "learning_rate": 0.000172479278722912,
1604
+ "loss": 0.6971,
1605
+ "step": 250
1606
+ },
1607
+ {
1608
+ "epoch": 0.75,
1609
+ "learning_rate": 0.00017226072475678782,
1610
+ "loss": 0.6289,
1611
+ "step": 251
1612
+ },
1613
+ {
1614
+ "epoch": 0.75,
1615
+ "learning_rate": 0.0001720414460571392,
1616
+ "loss": 0.6786,
1617
+ "step": 252
1618
+ },
1619
+ {
1620
+ "epoch": 0.76,
1621
+ "learning_rate": 0.0001718214448232054,
1622
+ "loss": 0.6599,
1623
+ "step": 253
1624
+ },
1625
+ {
1626
+ "epoch": 0.76,
1627
+ "learning_rate": 0.0001716007232614723,
1628
+ "loss": 0.656,
1629
+ "step": 254
1630
+ },
1631
+ {
1632
+ "epoch": 0.76,
1633
+ "learning_rate": 0.00017137928358565028,
1634
+ "loss": 0.7092,
1635
+ "step": 255
1636
+ },
1637
+ {
1638
+ "epoch": 0.76,
1639
+ "learning_rate": 0.000171157128016652,
1640
+ "loss": 0.6596,
1641
+ "step": 256
1642
+ },
1643
+ {
1644
+ "epoch": 0.77,
1645
+ "learning_rate": 0.00017093425878257005,
1646
+ "loss": 0.6937,
1647
+ "step": 257
1648
+ },
1649
+ {
1650
+ "epoch": 0.77,
1651
+ "learning_rate": 0.00017071067811865476,
1652
+ "loss": 0.7027,
1653
+ "step": 258
1654
+ },
1655
+ {
1656
+ "epoch": 0.77,
1657
+ "learning_rate": 0.00017048638826729163,
1658
+ "loss": 0.6776,
1659
+ "step": 259
1660
+ },
1661
+ {
1662
+ "epoch": 0.78,
1663
+ "learning_rate": 0.0001702613914779789,
1664
+ "loss": 0.7012,
1665
+ "step": 260
1666
+ },
1667
+ {
1668
+ "epoch": 0.78,
1669
+ "eval_loss": 0.9995876550674438,
1670
+ "eval_runtime": 15.664,
1671
+ "eval_samples_per_second": 37.219,
1672
+ "eval_steps_per_second": 18.641,
1673
+ "step": 260
1674
+ },
1675
+ {
1676
+ "epoch": 0.78,
1677
+ "learning_rate": 0.00017003569000730498,
1678
+ "loss": 0.6786,
1679
+ "step": 261
1680
+ },
1681
+ {
1682
+ "epoch": 0.78,
1683
+ "learning_rate": 0.0001698092861189259,
1684
+ "loss": 0.7071,
1685
+ "step": 262
1686
+ },
1687
+ {
1688
+ "epoch": 0.79,
1689
+ "learning_rate": 0.00016958218208354252,
1690
+ "loss": 0.6556,
1691
+ "step": 263
1692
+ },
1693
+ {
1694
+ "epoch": 0.79,
1695
+ "learning_rate": 0.00016935438017887772,
1696
+ "loss": 0.6418,
1697
+ "step": 264
1698
+ },
1699
+ {
1700
+ "epoch": 0.79,
1701
+ "learning_rate": 0.0001691258826896536,
1702
+ "loss": 0.6726,
1703
+ "step": 265
1704
+ },
1705
+ {
1706
+ "epoch": 0.79,
1707
+ "learning_rate": 0.00016889669190756868,
1708
+ "loss": 0.6514,
1709
+ "step": 266
1710
+ },
1711
+ {
1712
+ "epoch": 0.8,
1713
+ "learning_rate": 0.00016866681013127466,
1714
+ "loss": 0.6914,
1715
+ "step": 267
1716
+ },
1717
+ {
1718
+ "epoch": 0.8,
1719
+ "learning_rate": 0.00016843623966635366,
1720
+ "loss": 0.6884,
1721
+ "step": 268
1722
+ },
1723
+ {
1724
+ "epoch": 0.8,
1725
+ "learning_rate": 0.00016820498282529479,
1726
+ "loss": 0.6624,
1727
+ "step": 269
1728
+ },
1729
+ {
1730
+ "epoch": 0.81,
1731
+ "learning_rate": 0.0001679730419274713,
1732
+ "loss": 0.7241,
1733
+ "step": 270
1734
+ },
1735
+ {
1736
+ "epoch": 0.81,
1737
+ "learning_rate": 0.00016774041929911698,
1738
+ "loss": 0.6923,
1739
+ "step": 271
1740
+ },
1741
+ {
1742
+ "epoch": 0.81,
1743
+ "learning_rate": 0.0001675071172733031,
1744
+ "loss": 0.6817,
1745
+ "step": 272
1746
+ },
1747
+ {
1748
+ "epoch": 0.82,
1749
+ "learning_rate": 0.00016727313818991487,
1750
+ "loss": 0.7042,
1751
+ "step": 273
1752
+ },
1753
+ {
1754
+ "epoch": 0.82,
1755
+ "learning_rate": 0.00016703848439562785,
1756
+ "loss": 0.66,
1757
+ "step": 274
1758
+ },
1759
+ {
1760
+ "epoch": 0.82,
1761
+ "learning_rate": 0.00016680315824388483,
1762
+ "loss": 0.662,
1763
+ "step": 275
1764
+ },
1765
+ {
1766
+ "epoch": 0.82,
1767
+ "learning_rate": 0.00016656716209487174,
1768
+ "loss": 0.6174,
1769
+ "step": 276
1770
+ },
1771
+ {
1772
+ "epoch": 0.83,
1773
+ "learning_rate": 0.00016633049831549424,
1774
+ "loss": 0.6684,
1775
+ "step": 277
1776
+ },
1777
+ {
1778
+ "epoch": 0.83,
1779
+ "learning_rate": 0.0001660931692793541,
1780
+ "loss": 0.6581,
1781
+ "step": 278
1782
+ },
1783
+ {
1784
+ "epoch": 0.83,
1785
+ "learning_rate": 0.00016585517736672498,
1786
+ "loss": 0.6835,
1787
+ "step": 279
1788
+ },
1789
+ {
1790
+ "epoch": 0.84,
1791
+ "learning_rate": 0.000165616524964529,
1792
+ "loss": 0.6599,
1793
+ "step": 280
1794
+ },
1795
+ {
1796
+ "epoch": 0.84,
1797
+ "eval_loss": 0.9926242828369141,
1798
+ "eval_runtime": 15.669,
1799
+ "eval_samples_per_second": 37.207,
1800
+ "eval_steps_per_second": 18.636,
1801
+ "step": 280
1802
+ },
1803
+ {
1804
+ "epoch": 0.84,
1805
+ "learning_rate": 0.00016537721446631252,
1806
+ "loss": 0.6619,
1807
+ "step": 281
1808
+ },
1809
+ {
1810
+ "epoch": 0.84,
1811
+ "learning_rate": 0.00016513724827222227,
1812
+ "loss": 0.6557,
1813
+ "step": 282
1814
+ },
1815
+ {
1816
+ "epoch": 0.85,
1817
+ "learning_rate": 0.00016489662878898113,
1818
+ "loss": 0.6577,
1819
+ "step": 283
1820
+ },
1821
+ {
1822
+ "epoch": 0.85,
1823
+ "learning_rate": 0.00016465535842986434,
1824
+ "loss": 0.6423,
1825
+ "step": 284
1826
+ },
1827
+ {
1828
+ "epoch": 0.85,
1829
+ "learning_rate": 0.00016441343961467471,
1830
+ "loss": 0.6472,
1831
+ "step": 285
1832
+ },
1833
+ {
1834
+ "epoch": 0.85,
1835
+ "learning_rate": 0.000164170874769719,
1836
+ "loss": 0.6434,
1837
+ "step": 286
1838
+ },
1839
+ {
1840
+ "epoch": 0.86,
1841
+ "learning_rate": 0.00016392766632778312,
1842
+ "loss": 0.6461,
1843
+ "step": 287
1844
+ },
1845
+ {
1846
+ "epoch": 0.86,
1847
+ "learning_rate": 0.00016368381672810786,
1848
+ "loss": 0.6857,
1849
+ "step": 288
1850
+ },
1851
+ {
1852
+ "epoch": 0.86,
1853
+ "learning_rate": 0.00016343932841636456,
1854
+ "loss": 0.6927,
1855
+ "step": 289
1856
+ },
1857
+ {
1858
+ "epoch": 0.87,
1859
+ "learning_rate": 0.0001631942038446304,
1860
+ "loss": 0.6263,
1861
+ "step": 290
1862
+ },
1863
+ {
1864
+ "epoch": 0.87,
1865
+ "learning_rate": 0.00016294844547136395,
1866
+ "loss": 0.6857,
1867
+ "step": 291
1868
+ },
1869
+ {
1870
+ "epoch": 0.87,
1871
+ "learning_rate": 0.00016270205576138032,
1872
+ "loss": 0.682,
1873
+ "step": 292
1874
+ },
1875
+ {
1876
+ "epoch": 0.88,
1877
+ "learning_rate": 0.0001624550371858267,
1878
+ "loss": 0.6964,
1879
+ "step": 293
1880
+ },
1881
+ {
1882
+ "epoch": 0.88,
1883
+ "learning_rate": 0.00016220739222215738,
1884
+ "loss": 0.6261,
1885
+ "step": 294
1886
+ },
1887
+ {
1888
+ "epoch": 0.88,
1889
+ "learning_rate": 0.00016195912335410895,
1890
+ "loss": 0.6848,
1891
+ "step": 295
1892
+ },
1893
+ {
1894
+ "epoch": 0.88,
1895
+ "learning_rate": 0.00016171023307167545,
1896
+ "loss": 0.635,
1897
+ "step": 296
1898
+ },
1899
+ {
1900
+ "epoch": 0.89,
1901
+ "learning_rate": 0.0001614607238710833,
1902
+ "loss": 0.6941,
1903
+ "step": 297
1904
+ },
1905
+ {
1906
+ "epoch": 0.89,
1907
+ "learning_rate": 0.0001612105982547663,
1908
+ "loss": 0.6339,
1909
+ "step": 298
1910
+ },
1911
+ {
1912
+ "epoch": 0.89,
1913
+ "learning_rate": 0.0001609598587313406,
1914
+ "loss": 0.6298,
1915
+ "step": 299
1916
+ },
1917
+ {
1918
+ "epoch": 0.9,
1919
+ "learning_rate": 0.00016070850781557948,
1920
+ "loss": 0.6401,
1921
+ "step": 300
1922
+ },
1923
+ {
1924
+ "epoch": 0.9,
1925
+ "eval_loss": 0.9913042187690735,
1926
+ "eval_runtime": 15.6715,
1927
+ "eval_samples_per_second": 37.201,
1928
+ "eval_steps_per_second": 18.633,
1929
+ "step": 300
1930
+ },
1931
+ {
1932
+ "epoch": 0.9,
1933
+ "learning_rate": 0.00016045654802838812,
1934
+ "loss": 0.6483,
1935
+ "step": 301
1936
+ },
1937
+ {
1938
+ "epoch": 0.9,
1939
+ "learning_rate": 0.0001602039818967783,
1940
+ "loss": 0.6352,
1941
+ "step": 302
1942
+ },
1943
+ {
1944
+ "epoch": 0.91,
1945
+ "learning_rate": 0.00015995081195384315,
1946
+ "loss": 0.6344,
1947
+ "step": 303
1948
+ },
1949
+ {
1950
+ "epoch": 0.91,
1951
+ "learning_rate": 0.00015969704073873157,
1952
+ "loss": 0.6317,
1953
+ "step": 304
1954
+ },
1955
+ {
1956
+ "epoch": 0.91,
1957
+ "learning_rate": 0.000159442670796623,
1958
+ "loss": 0.6979,
1959
+ "step": 305
1960
+ },
1961
+ {
1962
+ "epoch": 0.91,
1963
+ "learning_rate": 0.0001591877046787017,
1964
+ "loss": 0.6429,
1965
+ "step": 306
1966
+ },
1967
+ {
1968
+ "epoch": 0.92,
1969
+ "learning_rate": 0.00015893214494213132,
1970
+ "loss": 0.6493,
1971
+ "step": 307
1972
+ },
1973
+ {
1974
+ "epoch": 0.92,
1975
+ "learning_rate": 0.00015867599415002895,
1976
+ "loss": 0.6765,
1977
+ "step": 308
1978
+ },
1979
+ {
1980
+ "epoch": 0.92,
1981
+ "learning_rate": 0.00015841925487143987,
1982
+ "loss": 0.6871,
1983
+ "step": 309
1984
+ },
1985
+ {
1986
+ "epoch": 0.93,
1987
+ "learning_rate": 0.00015816192968131138,
1988
+ "loss": 0.6389,
1989
+ "step": 310
1990
+ },
1991
+ {
1992
+ "epoch": 0.93,
1993
+ "learning_rate": 0.00015790402116046726,
1994
+ "loss": 0.6002,
1995
+ "step": 311
1996
+ },
1997
+ {
1998
+ "epoch": 0.93,
1999
+ "learning_rate": 0.0001576455318955816,
2000
+ "loss": 0.6025,
2001
+ "step": 312
2002
+ },
2003
+ {
2004
+ "epoch": 0.94,
2005
+ "learning_rate": 0.0001573864644791532,
2006
+ "loss": 0.6692,
2007
+ "step": 313
2008
+ },
2009
+ {
2010
+ "epoch": 0.94,
2011
+ "learning_rate": 0.00015712682150947923,
2012
+ "loss": 0.7079,
2013
+ "step": 314
2014
+ },
2015
+ {
2016
+ "epoch": 0.94,
2017
+ "learning_rate": 0.00015686660559062945,
2018
+ "loss": 0.6622,
2019
+ "step": 315
2020
+ },
2021
+ {
2022
+ "epoch": 0.94,
2023
+ "learning_rate": 0.00015660581933241993,
2024
+ "loss": 0.6402,
2025
+ "step": 316
2026
+ },
2027
+ {
2028
+ "epoch": 0.95,
2029
+ "learning_rate": 0.00015634446535038688,
2030
+ "loss": 0.7251,
2031
+ "step": 317
2032
+ },
2033
+ {
2034
+ "epoch": 0.95,
2035
+ "learning_rate": 0.00015608254626576048,
2036
+ "loss": 0.6806,
2037
+ "step": 318
2038
+ },
2039
+ {
2040
+ "epoch": 0.95,
2041
+ "learning_rate": 0.00015582006470543855,
2042
+ "loss": 0.6475,
2043
+ "step": 319
2044
+ },
2045
+ {
2046
+ "epoch": 0.96,
2047
+ "learning_rate": 0.00015555702330196023,
2048
+ "loss": 0.6665,
2049
+ "step": 320
2050
+ },
2051
+ {
2052
+ "epoch": 0.96,
2053
+ "eval_loss": 0.9910463094711304,
2054
+ "eval_runtime": 15.7019,
2055
+ "eval_samples_per_second": 37.129,
2056
+ "eval_steps_per_second": 18.596,
2057
+ "step": 320
2058
+ },
2059
+ {
2060
+ "epoch": 0.96,
2061
+ "learning_rate": 0.00015529342469347958,
2062
+ "loss": 0.6674,
2063
+ "step": 321
2064
+ },
2065
+ {
2066
+ "epoch": 0.96,
2067
+ "learning_rate": 0.00015502927152373914,
2068
+ "loss": 0.6873,
2069
+ "step": 322
2070
+ },
2071
+ {
2072
+ "epoch": 0.96,
2073
+ "learning_rate": 0.00015476456644204328,
2074
+ "loss": 0.6425,
2075
+ "step": 323
2076
+ },
2077
+ {
2078
+ "epoch": 0.97,
2079
+ "learning_rate": 0.0001544993121032318,
2080
+ "loss": 0.6565,
2081
+ "step": 324
2082
+ },
2083
+ {
2084
+ "epoch": 0.97,
2085
+ "learning_rate": 0.00015423351116765317,
2086
+ "loss": 0.6886,
2087
+ "step": 325
2088
+ },
2089
+ {
2090
+ "epoch": 0.97,
2091
+ "learning_rate": 0.000153967166301138,
2092
+ "loss": 0.6904,
2093
+ "step": 326
2094
+ },
2095
+ {
2096
+ "epoch": 0.98,
2097
+ "learning_rate": 0.00015370028017497217,
2098
+ "loss": 0.6281,
2099
+ "step": 327
2100
+ },
2101
+ {
2102
+ "epoch": 0.98,
2103
+ "learning_rate": 0.00015343285546587013,
2104
+ "loss": 0.6836,
2105
+ "step": 328
2106
+ },
2107
+ {
2108
+ "epoch": 0.98,
2109
+ "learning_rate": 0.00015316489485594782,
2110
+ "loss": 0.697,
2111
+ "step": 329
2112
+ },
2113
+ {
2114
+ "epoch": 0.99,
2115
+ "learning_rate": 0.00015289640103269625,
2116
+ "loss": 0.656,
2117
+ "step": 330
2118
+ },
2119
+ {
2120
+ "epoch": 0.99,
2121
+ "learning_rate": 0.00015262737668895405,
2122
+ "loss": 0.6649,
2123
+ "step": 331
2124
+ },
2125
+ {
2126
+ "epoch": 0.99,
2127
+ "learning_rate": 0.00015235782452288068,
2128
+ "loss": 0.6775,
2129
+ "step": 332
2130
+ },
2131
+ {
2132
+ "epoch": 0.99,
2133
+ "learning_rate": 0.00015208774723792943,
2134
+ "loss": 0.6637,
2135
+ "step": 333
2136
+ },
2137
+ {
2138
+ "epoch": 1.0,
2139
+ "learning_rate": 0.0001518171475428202,
2140
+ "loss": 0.7175,
2141
+ "step": 334
2142
+ },
2143
+ {
2144
+ "epoch": 1.0,
2145
+ "learning_rate": 0.00015154602815151233,
2146
+ "loss": 0.6345,
2147
+ "step": 335
2148
+ },
2149
+ {
2150
+ "epoch": 1.0,
2151
+ "learning_rate": 0.00015127439178317745,
2152
+ "loss": 0.6214,
2153
+ "step": 336
2154
+ },
2155
+ {
2156
+ "epoch": 1.01,
2157
+ "learning_rate": 0.00015100224116217217,
2158
+ "loss": 0.6026,
2159
+ "step": 337
2160
+ },
2161
+ {
2162
+ "epoch": 1.01,
2163
+ "learning_rate": 0.00015072957901801076,
2164
+ "loss": 0.6383,
2165
+ "step": 338
2166
+ },
2167
+ {
2168
+ "epoch": 1.01,
2169
+ "learning_rate": 0.00015045640808533772,
2170
+ "loss": 0.6026,
2171
+ "step": 339
2172
+ },
2173
+ {
2174
+ "epoch": 1.02,
2175
+ "learning_rate": 0.0001501827311039005,
2176
+ "loss": 0.5771,
2177
+ "step": 340
2178
+ },
2179
+ {
2180
+ "epoch": 1.02,
2181
+ "eval_loss": 0.9907026886940002,
2182
+ "eval_runtime": 15.6789,
2183
+ "eval_samples_per_second": 37.184,
2184
+ "eval_steps_per_second": 18.624,
2185
+ "step": 340
2186
+ },
2187
+ {
2188
+ "epoch": 1.02,
2189
+ "learning_rate": 0.00014990855081852185,
2190
+ "loss": 0.6087,
2191
+ "step": 341
2192
+ },
2193
+ {
2194
+ "epoch": 1.02,
2195
+ "learning_rate": 0.0001496338699790724,
2196
+ "loss": 0.6411,
2197
+ "step": 342
2198
+ },
2199
+ {
2200
+ "epoch": 1.02,
2201
+ "learning_rate": 0.00014935869134044308,
2202
+ "loss": 0.6074,
2203
+ "step": 343
2204
+ },
2205
+ {
2206
+ "epoch": 1.03,
2207
+ "learning_rate": 0.00014908301766251739,
2208
+ "loss": 0.5877,
2209
+ "step": 344
2210
+ },
2211
+ {
2212
+ "epoch": 1.03,
2213
+ "learning_rate": 0.00014880685171014382,
2214
+ "loss": 0.5731,
2215
+ "step": 345
2216
+ },
2217
+ {
2218
+ "epoch": 1.03,
2219
+ "learning_rate": 0.00014853019625310813,
2220
+ "loss": 0.6113,
2221
+ "step": 346
2222
+ },
2223
+ {
2224
+ "epoch": 1.04,
2225
+ "learning_rate": 0.00014825305406610545,
2226
+ "loss": 0.5965,
2227
+ "step": 347
2228
+ },
2229
+ {
2230
+ "epoch": 1.04,
2231
+ "learning_rate": 0.00014797542792871265,
2232
+ "loss": 0.6175,
2233
+ "step": 348
2234
+ },
2235
+ {
2236
+ "epoch": 1.04,
2237
+ "learning_rate": 0.0001476973206253602,
2238
+ "loss": 0.6018,
2239
+ "step": 349
2240
+ },
2241
+ {
2242
+ "epoch": 1.05,
2243
+ "learning_rate": 0.0001474187349453045,
2244
+ "loss": 0.6039,
2245
+ "step": 350
2246
+ },
2247
+ {
2248
+ "epoch": 1.05,
2249
+ "learning_rate": 0.0001471396736825998,
2250
+ "loss": 0.5692,
2251
+ "step": 351
2252
+ },
2253
+ {
2254
+ "epoch": 1.05,
2255
+ "learning_rate": 0.00014686013963607,
2256
+ "loss": 0.6234,
2257
+ "step": 352
2258
+ },
2259
+ {
2260
+ "epoch": 1.05,
2261
+ "learning_rate": 0.00014658013560928096,
2262
+ "loss": 0.5961,
2263
+ "step": 353
2264
+ },
2265
+ {
2266
+ "epoch": 1.06,
2267
+ "learning_rate": 0.00014629966441051208,
2268
+ "loss": 0.6315,
2269
+ "step": 354
2270
+ },
2271
+ {
2272
+ "epoch": 1.06,
2273
+ "learning_rate": 0.00014601872885272824,
2274
+ "loss": 0.6223,
2275
+ "step": 355
2276
+ },
2277
+ {
2278
+ "epoch": 1.06,
2279
+ "learning_rate": 0.0001457373317535515,
2280
+ "loss": 0.6557,
2281
+ "step": 356
2282
+ },
2283
+ {
2284
+ "epoch": 1.07,
2285
+ "learning_rate": 0.00014545547593523307,
2286
+ "loss": 0.6012,
2287
+ "step": 357
2288
+ },
2289
+ {
2290
+ "epoch": 1.07,
2291
+ "learning_rate": 0.0001451731642246247,
2292
+ "loss": 0.6405,
2293
+ "step": 358
2294
+ },
2295
+ {
2296
+ "epoch": 1.07,
2297
+ "learning_rate": 0.0001448903994531506,
2298
+ "loss": 0.6106,
2299
+ "step": 359
2300
+ },
2301
+ {
2302
+ "epoch": 1.08,
2303
+ "learning_rate": 0.00014460718445677876,
2304
+ "loss": 0.6286,
2305
+ "step": 360
2306
+ },
2307
+ {
2308
+ "epoch": 1.08,
2309
+ "eval_loss": 0.9830257892608643,
2310
+ "eval_runtime": 15.7168,
2311
+ "eval_samples_per_second": 37.094,
2312
+ "eval_steps_per_second": 18.579,
2313
+ "step": 360
2314
+ },
2315
+ {
2316
+ "epoch": 1.08,
2317
+ "learning_rate": 0.00014432352207599286,
2318
+ "loss": 0.626,
2319
+ "step": 361
2320
+ },
2321
+ {
2322
+ "epoch": 1.08,
2323
+ "learning_rate": 0.00014403941515576344,
2324
+ "loss": 0.6169,
2325
+ "step": 362
2326
+ },
2327
+ {
2328
+ "epoch": 1.08,
2329
+ "learning_rate": 0.0001437548665455196,
2330
+ "loss": 0.572,
2331
+ "step": 363
2332
+ },
2333
+ {
2334
+ "epoch": 1.09,
2335
+ "learning_rate": 0.00014346987909912023,
2336
+ "loss": 0.6255,
2337
+ "step": 364
2338
+ },
2339
+ {
2340
+ "epoch": 1.09,
2341
+ "learning_rate": 0.0001431844556748257,
2342
+ "loss": 0.6036,
2343
+ "step": 365
2344
+ },
2345
+ {
2346
+ "epoch": 1.09,
2347
+ "learning_rate": 0.00014289859913526874,
2348
+ "loss": 0.5804,
2349
+ "step": 366
2350
+ },
2351
+ {
2352
+ "epoch": 1.1,
2353
+ "learning_rate": 0.00014261231234742619,
2354
+ "loss": 0.5898,
2355
+ "step": 367
2356
+ },
2357
+ {
2358
+ "epoch": 1.1,
2359
+ "learning_rate": 0.00014232559818258984,
2360
+ "loss": 0.6358,
2361
+ "step": 368
2362
+ },
2363
+ {
2364
+ "epoch": 1.1,
2365
+ "learning_rate": 0.000142038459516338,
2366
+ "loss": 0.5652,
2367
+ "step": 369
2368
+ },
2369
+ {
2370
+ "epoch": 1.11,
2371
+ "learning_rate": 0.00014175089922850633,
2372
+ "loss": 0.6259,
2373
+ "step": 370
2374
+ },
2375
+ {
2376
+ "epoch": 1.11,
2377
+ "learning_rate": 0.00014146292020315924,
2378
+ "loss": 0.6309,
2379
+ "step": 371
2380
+ },
2381
+ {
2382
+ "epoch": 1.11,
2383
+ "learning_rate": 0.00014117452532856083,
2384
+ "loss": 0.6325,
2385
+ "step": 372
2386
+ },
2387
+ {
2388
+ "epoch": 1.11,
2389
+ "learning_rate": 0.0001408857174971458,
2390
+ "loss": 0.5984,
2391
+ "step": 373
2392
+ },
2393
+ {
2394
+ "epoch": 1.12,
2395
+ "learning_rate": 0.0001405964996054907,
2396
+ "loss": 0.5981,
2397
+ "step": 374
2398
+ },
2399
+ {
2400
+ "epoch": 1.12,
2401
+ "learning_rate": 0.00014030687455428474,
2402
+ "loss": 0.6407,
2403
+ "step": 375
2404
+ },
2405
+ {
2406
+ "epoch": 1.12,
2407
+ "learning_rate": 0.00014001684524830057,
2408
+ "loss": 0.5686,
2409
+ "step": 376
2410
+ },
2411
+ {
2412
+ "epoch": 1.13,
2413
+ "learning_rate": 0.0001397264145963655,
2414
+ "loss": 0.6151,
2415
+ "step": 377
2416
+ },
2417
+ {
2418
+ "epoch": 1.13,
2419
+ "learning_rate": 0.00013943558551133186,
2420
+ "loss": 0.5925,
2421
+ "step": 378
2422
+ },
2423
+ {
2424
+ "epoch": 1.13,
2425
+ "learning_rate": 0.00013914436091004827,
2426
+ "loss": 0.5872,
2427
+ "step": 379
2428
+ },
2429
+ {
2430
+ "epoch": 1.14,
2431
+ "learning_rate": 0.00013885274371333,
2432
+ "loss": 0.6064,
2433
+ "step": 380
2434
+ },
2435
+ {
2436
+ "epoch": 1.14,
2437
+ "eval_loss": 0.9865307807922363,
2438
+ "eval_runtime": 15.7209,
2439
+ "eval_samples_per_second": 37.084,
2440
+ "eval_steps_per_second": 18.574,
2441
+ "step": 380
2442
+ },
2443
+ {
2444
+ "epoch": 1.14,
2445
+ "learning_rate": 0.00013856073684592995,
2446
+ "loss": 0.6151,
2447
+ "step": 381
2448
+ },
2449
+ {
2450
+ "epoch": 1.14,
2451
+ "learning_rate": 0.000138268343236509,
2452
+ "loss": 0.6262,
2453
+ "step": 382
2454
+ },
2455
+ {
2456
+ "epoch": 1.14,
2457
+ "learning_rate": 0.000137975565817607,
2458
+ "loss": 0.5727,
2459
+ "step": 383
2460
+ },
2461
+ {
2462
+ "epoch": 1.15,
2463
+ "learning_rate": 0.00013768240752561314,
2464
+ "loss": 0.6486,
2465
+ "step": 384
2466
+ },
2467
+ {
2468
+ "epoch": 1.15,
2469
+ "learning_rate": 0.00013738887130073654,
2470
+ "loss": 0.6102,
2471
+ "step": 385
2472
+ },
2473
+ {
2474
+ "epoch": 1.15,
2475
+ "learning_rate": 0.0001370949600869768,
2476
+ "loss": 0.591,
2477
+ "step": 386
2478
+ },
2479
+ {
2480
+ "epoch": 1.16,
2481
+ "learning_rate": 0.00013680067683209438,
2482
+ "loss": 0.6336,
2483
+ "step": 387
2484
+ },
2485
+ {
2486
+ "epoch": 1.16,
2487
+ "learning_rate": 0.00013650602448758112,
2488
+ "loss": 0.6699,
2489
+ "step": 388
2490
+ },
2491
+ {
2492
+ "epoch": 1.16,
2493
+ "learning_rate": 0.00013621100600863065,
2494
+ "loss": 0.6317,
2495
+ "step": 389
2496
+ },
2497
+ {
2498
+ "epoch": 1.17,
2499
+ "learning_rate": 0.0001359156243541087,
2500
+ "loss": 0.6129,
2501
+ "step": 390
2502
+ },
2503
+ {
2504
+ "epoch": 1.17,
2505
+ "learning_rate": 0.00013561988248652346,
2506
+ "loss": 0.6984,
2507
+ "step": 391
2508
+ },
2509
+ {
2510
+ "epoch": 1.17,
2511
+ "learning_rate": 0.00013532378337199582,
2512
+ "loss": 0.5925,
2513
+ "step": 392
2514
+ },
2515
+ {
2516
+ "epoch": 1.17,
2517
+ "learning_rate": 0.00013502732998022966,
2518
+ "loss": 0.5835,
2519
+ "step": 393
2520
+ },
2521
+ {
2522
+ "epoch": 1.18,
2523
+ "learning_rate": 0.00013473052528448201,
2524
+ "loss": 0.5919,
2525
+ "step": 394
2526
+ },
2527
+ {
2528
+ "epoch": 1.18,
2529
+ "learning_rate": 0.00013443337226153344,
2530
+ "loss": 0.621,
2531
+ "step": 395
2532
+ },
2533
+ {
2534
+ "epoch": 1.18,
2535
+ "learning_rate": 0.00013413587389165784,
2536
+ "loss": 0.6363,
2537
+ "step": 396
2538
+ },
2539
+ {
2540
+ "epoch": 1.19,
2541
+ "learning_rate": 0.0001338380331585928,
2542
+ "loss": 0.6383,
2543
+ "step": 397
2544
+ },
2545
+ {
2546
+ "epoch": 1.19,
2547
+ "learning_rate": 0.00013353985304950973,
2548
+ "loss": 0.5803,
2549
+ "step": 398
2550
+ },
2551
+ {
2552
+ "epoch": 1.19,
2553
+ "learning_rate": 0.0001332413365549836,
2554
+ "loss": 0.6378,
2555
+ "step": 399
2556
+ },
2557
+ {
2558
+ "epoch": 1.19,
2559
+ "learning_rate": 0.00013294248666896328,
2560
+ "loss": 0.5976,
2561
+ "step": 400
2562
+ },
2563
+ {
2564
+ "epoch": 1.19,
2565
+ "eval_loss": 0.980188250541687,
2566
+ "eval_runtime": 15.7209,
2567
+ "eval_samples_per_second": 37.084,
2568
+ "eval_steps_per_second": 18.574,
2569
+ "step": 400
2570
+ },
2571
+ {
2572
+ "epoch": 1.2,
2573
+ "learning_rate": 0.00013264330638874128,
2574
+ "loss": 0.6282,
2575
+ "step": 401
2576
+ },
2577
+ {
2578
+ "epoch": 1.2,
2579
+ "learning_rate": 0.0001323437987149238,
2580
+ "loss": 0.6146,
2581
+ "step": 402
2582
+ },
2583
+ {
2584
+ "epoch": 1.2,
2585
+ "learning_rate": 0.00013204396665140067,
2586
+ "loss": 0.6089,
2587
+ "step": 403
2588
+ },
2589
+ {
2590
+ "epoch": 1.21,
2591
+ "learning_rate": 0.00013174381320531505,
2592
+ "loss": 0.5657,
2593
+ "step": 404
2594
+ },
2595
+ {
2596
+ "epoch": 1.21,
2597
+ "learning_rate": 0.0001314433413870334,
2598
+ "loss": 0.6638,
2599
+ "step": 405
2600
+ },
2601
+ {
2602
+ "epoch": 1.21,
2603
+ "learning_rate": 0.0001311425542101154,
2604
+ "loss": 0.6129,
2605
+ "step": 406
2606
+ },
2607
+ {
2608
+ "epoch": 1.22,
2609
+ "learning_rate": 0.0001308414546912834,
2610
+ "loss": 0.5862,
2611
+ "step": 407
2612
+ },
2613
+ {
2614
+ "epoch": 1.22,
2615
+ "learning_rate": 0.00013054004585039258,
2616
+ "loss": 0.5759,
2617
+ "step": 408
2618
+ },
2619
+ {
2620
+ "epoch": 1.22,
2621
+ "learning_rate": 0.00013023833071040025,
2622
+ "loss": 0.6275,
2623
+ "step": 409
2624
+ },
2625
+ {
2626
+ "epoch": 1.22,
2627
+ "learning_rate": 0.00012993631229733582,
2628
+ "loss": 0.6095,
2629
+ "step": 410
2630
+ },
2631
+ {
2632
+ "epoch": 1.23,
2633
+ "learning_rate": 0.0001296339936402703,
2634
+ "loss": 0.6187,
2635
+ "step": 411
2636
+ },
2637
+ {
2638
+ "epoch": 1.23,
2639
+ "learning_rate": 0.00012933137777128607,
2640
+ "loss": 0.603,
2641
+ "step": 412
2642
+ },
2643
+ {
2644
+ "epoch": 1.23,
2645
+ "learning_rate": 0.00012902846772544624,
2646
+ "loss": 0.5911,
2647
+ "step": 413
2648
+ },
2649
+ {
2650
+ "epoch": 1.24,
2651
+ "learning_rate": 0.0001287252665407645,
2652
+ "loss": 0.5979,
2653
+ "step": 414
2654
+ },
2655
+ {
2656
+ "epoch": 1.24,
2657
+ "learning_rate": 0.00012842177725817434,
2658
+ "loss": 0.6157,
2659
+ "step": 415
2660
+ },
2661
+ {
2662
+ "epoch": 1.24,
2663
+ "learning_rate": 0.0001281180029214988,
2664
+ "loss": 0.5796,
2665
+ "step": 416
2666
+ },
2667
+ {
2668
+ "epoch": 1.25,
2669
+ "learning_rate": 0.00012781394657741989,
2670
+ "loss": 0.6215,
2671
+ "step": 417
2672
+ },
2673
+ {
2674
+ "epoch": 1.25,
2675
+ "learning_rate": 0.0001275096112754478,
2676
+ "loss": 0.5927,
2677
+ "step": 418
2678
+ },
2679
+ {
2680
+ "epoch": 1.25,
2681
+ "learning_rate": 0.0001272050000678908,
2682
+ "loss": 0.628,
2683
+ "step": 419
2684
+ },
2685
+ {
2686
+ "epoch": 1.25,
2687
+ "learning_rate": 0.000126900116009824,
2688
+ "loss": 0.5512,
2689
+ "step": 420
2690
+ },
2691
+ {
2692
+ "epoch": 1.25,
2693
+ "eval_loss": 0.9817403554916382,
2694
+ "eval_runtime": 15.7224,
2695
+ "eval_samples_per_second": 37.081,
2696
+ "eval_steps_per_second": 18.572,
2697
+ "step": 420
2698
+ },
2699
+ {
2700
+ "epoch": 1.26,
2701
+ "learning_rate": 0.00012659496215905938,
2702
+ "loss": 0.6477,
2703
+ "step": 421
2704
+ },
2705
+ {
2706
+ "epoch": 1.26,
2707
+ "learning_rate": 0.0001262895415761145,
2708
+ "loss": 0.6304,
2709
+ "step": 422
2710
+ },
2711
+ {
2712
+ "epoch": 1.26,
2713
+ "learning_rate": 0.00012598385732418224,
2714
+ "loss": 0.6261,
2715
+ "step": 423
2716
+ },
2717
+ {
2718
+ "epoch": 1.27,
2719
+ "learning_rate": 0.00012567791246909994,
2720
+ "loss": 0.6472,
2721
+ "step": 424
2722
+ },
2723
+ {
2724
+ "epoch": 1.27,
2725
+ "learning_rate": 0.00012537171007931858,
2726
+ "loss": 0.6299,
2727
+ "step": 425
2728
+ },
2729
+ {
2730
+ "epoch": 1.27,
2731
+ "learning_rate": 0.00012506525322587207,
2732
+ "loss": 0.605,
2733
+ "step": 426
2734
+ },
2735
+ {
2736
+ "epoch": 1.28,
2737
+ "learning_rate": 0.00012475854498234648,
2738
+ "loss": 0.5852,
2739
+ "step": 427
2740
+ },
2741
+ {
2742
+ "epoch": 1.28,
2743
+ "learning_rate": 0.0001244515884248491,
2744
+ "loss": 0.5736,
2745
+ "step": 428
2746
+ },
2747
+ {
2748
+ "epoch": 1.28,
2749
+ "learning_rate": 0.0001241443866319778,
2750
+ "loss": 0.5795,
2751
+ "step": 429
2752
+ },
2753
+ {
2754
+ "epoch": 1.28,
2755
+ "learning_rate": 0.00012383694268478993,
2756
+ "loss": 0.6088,
2757
+ "step": 430
2758
+ },
2759
+ {
2760
+ "epoch": 1.29,
2761
+ "learning_rate": 0.00012352925966677147,
2762
+ "loss": 0.6547,
2763
+ "step": 431
2764
+ },
2765
+ {
2766
+ "epoch": 1.29,
2767
+ "learning_rate": 0.0001232213406638062,
2768
+ "loss": 0.5549,
2769
+ "step": 432
2770
+ },
2771
+ {
2772
+ "epoch": 1.29,
2773
+ "learning_rate": 0.00012291318876414477,
2774
+ "loss": 0.6398,
2775
+ "step": 433
2776
+ },
2777
+ {
2778
+ "epoch": 1.3,
2779
+ "learning_rate": 0.0001226048070583735,
2780
+ "loss": 0.6377,
2781
+ "step": 434
2782
+ },
2783
+ {
2784
+ "epoch": 1.3,
2785
+ "learning_rate": 0.0001222961986393836,
2786
+ "loss": 0.6207,
2787
+ "step": 435
2788
+ },
2789
+ {
2790
+ "epoch": 1.3,
2791
+ "learning_rate": 0.00012198736660234009,
2792
+ "loss": 0.5999,
2793
+ "step": 436
2794
+ },
2795
+ {
2796
+ "epoch": 1.31,
2797
+ "learning_rate": 0.00012167831404465078,
2798
+ "loss": 0.6088,
2799
+ "step": 437
2800
+ },
2801
+ {
2802
+ "epoch": 1.31,
2803
+ "learning_rate": 0.00012136904406593507,
2804
+ "loss": 0.6525,
2805
+ "step": 438
2806
+ },
2807
+ {
2808
+ "epoch": 1.31,
2809
+ "learning_rate": 0.00012105955976799314,
2810
+ "loss": 0.5816,
2811
+ "step": 439
2812
+ },
2813
+ {
2814
+ "epoch": 1.31,
2815
+ "learning_rate": 0.00012074986425477445,
2816
+ "loss": 0.6333,
2817
+ "step": 440
2818
+ },
2819
+ {
2820
+ "epoch": 1.31,
2821
+ "eval_loss": 0.9809613227844238,
2822
+ "eval_runtime": 15.7148,
2823
+ "eval_samples_per_second": 37.099,
2824
+ "eval_steps_per_second": 18.581,
2825
+ "step": 440
2826
+ },
2827
+ {
2828
+ "epoch": 1.32,
2829
+ "learning_rate": 0.00012043996063234706,
2830
+ "loss": 0.5735,
2831
+ "step": 441
2832
+ },
2833
+ {
2834
+ "epoch": 1.32,
2835
+ "learning_rate": 0.00012012985200886602,
2836
+ "loss": 0.5884,
2837
+ "step": 442
2838
+ },
2839
+ {
2840
+ "epoch": 1.32,
2841
+ "learning_rate": 0.00011981954149454262,
2842
+ "loss": 0.5983,
2843
+ "step": 443
2844
+ },
2845
+ {
2846
+ "epoch": 1.33,
2847
+ "learning_rate": 0.00011950903220161285,
2848
+ "loss": 0.6161,
2849
+ "step": 444
2850
+ },
2851
+ {
2852
+ "epoch": 1.33,
2853
+ "learning_rate": 0.00011919832724430642,
2854
+ "loss": 0.5827,
2855
+ "step": 445
2856
+ },
2857
+ {
2858
+ "epoch": 1.33,
2859
+ "learning_rate": 0.00011888742973881543,
2860
+ "loss": 0.5955,
2861
+ "step": 446
2862
+ },
2863
+ {
2864
+ "epoch": 1.34,
2865
+ "learning_rate": 0.00011857634280326311,
2866
+ "loss": 0.602,
2867
+ "step": 447
2868
+ },
2869
+ {
2870
+ "epoch": 1.34,
2871
+ "learning_rate": 0.00011826506955767258,
2872
+ "loss": 0.5978,
2873
+ "step": 448
2874
+ },
2875
+ {
2876
+ "epoch": 1.34,
2877
+ "learning_rate": 0.00011795361312393549,
2878
+ "loss": 0.6562,
2879
+ "step": 449
2880
+ },
2881
+ {
2882
+ "epoch": 1.34,
2883
+ "learning_rate": 0.00011764197662578086,
2884
+ "loss": 0.6053,
2885
+ "step": 450
2886
+ },
2887
+ {
2888
+ "epoch": 1.35,
2889
+ "learning_rate": 0.00011733016318874357,
2890
+ "loss": 0.6176,
2891
+ "step": 451
2892
+ },
2893
+ {
2894
+ "epoch": 1.35,
2895
+ "learning_rate": 0.00011701817594013312,
2896
+ "loss": 0.6388,
2897
+ "step": 452
2898
+ },
2899
+ {
2900
+ "epoch": 1.35,
2901
+ "learning_rate": 0.00011670601800900225,
2902
+ "loss": 0.599,
2903
+ "step": 453
2904
+ },
2905
+ {
2906
+ "epoch": 1.36,
2907
+ "learning_rate": 0.00011639369252611552,
2908
+ "loss": 0.6294,
2909
+ "step": 454
2910
+ },
2911
+ {
2912
+ "epoch": 1.36,
2913
+ "learning_rate": 0.00011608120262391801,
2914
+ "loss": 0.5844,
2915
+ "step": 455
2916
+ },
2917
+ {
2918
+ "epoch": 1.36,
2919
+ "learning_rate": 0.00011576855143650371,
2920
+ "loss": 0.6244,
2921
+ "step": 456
2922
+ },
2923
+ {
2924
+ "epoch": 1.37,
2925
+ "learning_rate": 0.00011545574209958432,
2926
+ "loss": 0.6068,
2927
+ "step": 457
2928
+ },
2929
+ {
2930
+ "epoch": 1.37,
2931
+ "learning_rate": 0.00011514277775045768,
2932
+ "loss": 0.5585,
2933
+ "step": 458
2934
+ },
2935
+ {
2936
+ "epoch": 1.37,
2937
+ "learning_rate": 0.0001148296615279762,
2938
+ "loss": 0.5998,
2939
+ "step": 459
2940
+ },
2941
+ {
2942
+ "epoch": 1.37,
2943
+ "learning_rate": 0.00011451639657251563,
2944
+ "loss": 0.5883,
2945
+ "step": 460
2946
+ },
2947
+ {
2948
+ "epoch": 1.37,
2949
+ "eval_loss": 0.9817206859588623,
2950
+ "eval_runtime": 15.8012,
2951
+ "eval_samples_per_second": 36.896,
2952
+ "eval_steps_per_second": 18.48,
2953
+ "step": 460
2954
+ },
2955
+ {
2956
+ "epoch": 1.38,
2957
+ "learning_rate": 0.00011420298602594332,
2958
+ "loss": 0.6029,
2959
+ "step": 461
2960
+ },
2961
+ {
2962
+ "epoch": 1.38,
2963
+ "learning_rate": 0.00011388943303158693,
2964
+ "loss": 0.6066,
2965
+ "step": 462
2966
+ },
2967
+ {
2968
+ "epoch": 1.38,
2969
+ "learning_rate": 0.00011357574073420265,
2970
+ "loss": 0.6229,
2971
+ "step": 463
2972
+ },
2973
+ {
2974
+ "epoch": 1.39,
2975
+ "learning_rate": 0.00011326191227994391,
2976
+ "loss": 0.62,
2977
+ "step": 464
2978
+ },
2979
+ {
2980
+ "epoch": 1.39,
2981
+ "learning_rate": 0.00011294795081632967,
2982
+ "loss": 0.6268,
2983
+ "step": 465
2984
+ },
2985
+ {
2986
+ "epoch": 1.39,
2987
+ "learning_rate": 0.00011263385949221295,
2988
+ "loss": 0.6215,
2989
+ "step": 466
2990
+ },
2991
+ {
2992
+ "epoch": 1.4,
2993
+ "learning_rate": 0.00011231964145774905,
2994
+ "loss": 0.6612,
2995
+ "step": 467
2996
+ },
2997
+ {
2998
+ "epoch": 1.4,
2999
+ "learning_rate": 0.0001120052998643643,
3000
+ "loss": 0.6015,
3001
+ "step": 468
3002
+ },
3003
+ {
3004
+ "epoch": 1.4,
3005
+ "learning_rate": 0.00011169083786472408,
3006
+ "loss": 0.6247,
3007
+ "step": 469
3008
+ },
3009
+ {
3010
+ "epoch": 1.4,
3011
+ "learning_rate": 0.00011137625861270151,
3012
+ "loss": 0.6472,
3013
+ "step": 470
3014
+ },
3015
+ {
3016
+ "epoch": 1.41,
3017
+ "learning_rate": 0.00011106156526334558,
3018
+ "loss": 0.6283,
3019
+ "step": 471
3020
+ },
3021
+ {
3022
+ "epoch": 1.41,
3023
+ "learning_rate": 0.00011074676097284973,
3024
+ "loss": 0.6026,
3025
+ "step": 472
3026
+ },
3027
+ {
3028
+ "epoch": 1.41,
3029
+ "learning_rate": 0.00011043184889851992,
3030
+ "loss": 0.669,
3031
+ "step": 473
3032
+ },
3033
+ {
3034
+ "epoch": 1.42,
3035
+ "learning_rate": 0.00011011683219874323,
3036
+ "loss": 0.6302,
3037
+ "step": 474
3038
+ },
3039
+ {
3040
+ "epoch": 1.42,
3041
+ "learning_rate": 0.0001098017140329561,
3042
+ "loss": 0.6662,
3043
+ "step": 475
3044
+ },
3045
+ {
3046
+ "epoch": 1.42,
3047
+ "learning_rate": 0.00010948649756161246,
3048
+ "loss": 0.6014,
3049
+ "step": 476
3050
+ },
3051
+ {
3052
+ "epoch": 1.42,
3053
+ "learning_rate": 0.00010917118594615237,
3054
+ "loss": 0.5959,
3055
+ "step": 477
3056
+ },
3057
+ {
3058
+ "epoch": 1.43,
3059
+ "learning_rate": 0.00010885578234897003,
3060
+ "loss": 0.5983,
3061
+ "step": 478
3062
+ },
3063
+ {
3064
+ "epoch": 1.43,
3065
+ "learning_rate": 0.00010854028993338221,
3066
+ "loss": 0.5428,
3067
+ "step": 479
3068
+ },
3069
+ {
3070
+ "epoch": 1.43,
3071
+ "learning_rate": 0.00010822471186359639,
3072
+ "loss": 0.5822,
3073
+ "step": 480
3074
+ },
3075
+ {
3076
+ "epoch": 1.43,
3077
+ "eval_loss": 0.978273868560791,
3078
+ "eval_runtime": 15.7427,
3079
+ "eval_samples_per_second": 37.033,
3080
+ "eval_steps_per_second": 18.548,
3081
+ "step": 480
3082
+ },
3083
+ {
3084
+ "epoch": 1.44,
3085
+ "learning_rate": 0.00010790905130467923,
3086
+ "loss": 0.635,
3087
+ "step": 481
3088
+ },
3089
+ {
3090
+ "epoch": 1.44,
3091
+ "learning_rate": 0.00010759331142252462,
3092
+ "loss": 0.5787,
3093
+ "step": 482
3094
+ },
3095
+ {
3096
+ "epoch": 1.44,
3097
+ "learning_rate": 0.00010727749538382211,
3098
+ "loss": 0.617,
3099
+ "step": 483
3100
+ },
3101
+ {
3102
+ "epoch": 1.45,
3103
+ "learning_rate": 0.00010696160635602487,
3104
+ "loss": 0.5646,
3105
+ "step": 484
3106
+ },
3107
+ {
3108
+ "epoch": 1.45,
3109
+ "learning_rate": 0.00010664564750731839,
3110
+ "loss": 0.6083,
3111
+ "step": 485
3112
+ },
3113
+ {
3114
+ "epoch": 1.45,
3115
+ "learning_rate": 0.00010632962200658815,
3116
+ "loss": 0.6098,
3117
+ "step": 486
3118
+ },
3119
+ {
3120
+ "epoch": 1.45,
3121
+ "learning_rate": 0.0001060135330233883,
3122
+ "loss": 0.5669,
3123
+ "step": 487
3124
+ },
3125
+ {
3126
+ "epoch": 1.46,
3127
+ "learning_rate": 0.00010569738372790956,
3128
+ "loss": 0.6169,
3129
+ "step": 488
3130
+ },
3131
+ {
3132
+ "epoch": 1.46,
3133
+ "learning_rate": 0.00010538117729094766,
3134
+ "loss": 0.595,
3135
+ "step": 489
3136
+ },
3137
+ {
3138
+ "epoch": 1.46,
3139
+ "learning_rate": 0.00010506491688387127,
3140
+ "loss": 0.6104,
3141
+ "step": 490
3142
+ },
3143
+ {
3144
+ "epoch": 1.47,
3145
+ "learning_rate": 0.00010474860567859057,
3146
+ "loss": 0.5904,
3147
+ "step": 491
3148
+ },
3149
+ {
3150
+ "epoch": 1.47,
3151
+ "learning_rate": 0.000104432246847525,
3152
+ "loss": 0.5679,
3153
+ "step": 492
3154
+ },
3155
+ {
3156
+ "epoch": 1.47,
3157
+ "learning_rate": 0.00010411584356357178,
3158
+ "loss": 0.5984,
3159
+ "step": 493
3160
+ },
3161
+ {
3162
+ "epoch": 1.48,
3163
+ "learning_rate": 0.00010379939900007393,
3164
+ "loss": 0.6033,
3165
+ "step": 494
3166
+ },
3167
+ {
3168
+ "epoch": 1.48,
3169
+ "learning_rate": 0.0001034829163307885,
3170
+ "loss": 0.6357,
3171
+ "step": 495
3172
+ },
3173
+ {
3174
+ "epoch": 1.48,
3175
+ "learning_rate": 0.00010316639872985472,
3176
+ "loss": 0.6547,
3177
+ "step": 496
3178
+ },
3179
+ {
3180
+ "epoch": 1.48,
3181
+ "learning_rate": 0.00010284984937176212,
3182
+ "loss": 0.6082,
3183
+ "step": 497
3184
+ },
3185
+ {
3186
+ "epoch": 1.49,
3187
+ "learning_rate": 0.00010253327143131879,
3188
+ "loss": 0.6176,
3189
+ "step": 498
3190
+ },
3191
+ {
3192
+ "epoch": 1.49,
3193
+ "learning_rate": 0.00010221666808361947,
3194
+ "loss": 0.5942,
3195
+ "step": 499
3196
+ },
3197
+ {
3198
+ "epoch": 1.49,
3199
+ "learning_rate": 0.00010190004250401368,
3200
+ "loss": 0.5878,
3201
+ "step": 500
3202
+ },
3203
+ {
3204
+ "epoch": 1.49,
3205
+ "eval_loss": 0.9756883978843689,
3206
+ "eval_runtime": 15.7172,
3207
+ "eval_samples_per_second": 37.093,
3208
+ "eval_steps_per_second": 18.578,
3209
+ "step": 500
3210
+ },
3211
+ {
3212
+ "epoch": 1.5,
3213
+ "learning_rate": 0.00010158339786807401,
3214
+ "loss": 0.6009,
3215
+ "step": 501
3216
+ },
3217
+ {
3218
+ "epoch": 1.5,
3219
+ "learning_rate": 0.00010126673735156402,
3220
+ "loss": 0.6042,
3221
+ "step": 502
3222
+ },
3223
+ {
3224
+ "epoch": 1.5,
3225
+ "learning_rate": 0.0001009500641304067,
3226
+ "loss": 0.6045,
3227
+ "step": 503
3228
+ },
3229
+ {
3230
+ "epoch": 1.51,
3231
+ "learning_rate": 0.00010063338138065234,
3232
+ "loss": 0.5939,
3233
+ "step": 504
3234
+ },
3235
+ {
3236
+ "epoch": 1.51,
3237
+ "learning_rate": 0.00010031669227844694,
3238
+ "loss": 0.6093,
3239
+ "step": 505
3240
+ },
3241
+ {
3242
+ "epoch": 1.51,
3243
+ "learning_rate": 0.0001,
3244
+ "loss": 0.6265,
3245
+ "step": 506
3246
+ },
3247
+ {
3248
+ "epoch": 1.51,
3249
+ "learning_rate": 9.968330772155311e-05,
3250
+ "loss": 0.5842,
3251
+ "step": 507
3252
+ },
3253
+ {
3254
+ "epoch": 1.52,
3255
+ "learning_rate": 9.936661861934765e-05,
3256
+ "loss": 0.6032,
3257
+ "step": 508
3258
+ },
3259
+ {
3260
+ "epoch": 1.52,
3261
+ "learning_rate": 9.904993586959332e-05,
3262
+ "loss": 0.5963,
3263
+ "step": 509
3264
+ },
3265
+ {
3266
+ "epoch": 1.52,
3267
+ "learning_rate": 9.8733262648436e-05,
3268
+ "loss": 0.5725,
3269
+ "step": 510
3270
+ },
3271
+ {
3272
+ "epoch": 1.53,
3273
+ "learning_rate": 9.841660213192605e-05,
3274
+ "loss": 0.6323,
3275
+ "step": 511
3276
+ },
3277
+ {
3278
+ "epoch": 1.53,
3279
+ "learning_rate": 9.809995749598632e-05,
3280
+ "loss": 0.6186,
3281
+ "step": 512
3282
+ },
3283
+ {
3284
+ "epoch": 1.53,
3285
+ "learning_rate": 9.778333191638055e-05,
3286
+ "loss": 0.6437,
3287
+ "step": 513
3288
+ },
3289
+ {
3290
+ "epoch": 1.54,
3291
+ "learning_rate": 9.746672856868123e-05,
3292
+ "loss": 0.608,
3293
+ "step": 514
3294
+ },
3295
+ {
3296
+ "epoch": 1.54,
3297
+ "learning_rate": 9.715015062823789e-05,
3298
+ "loss": 0.6013,
3299
+ "step": 515
3300
+ },
3301
+ {
3302
+ "epoch": 1.54,
3303
+ "learning_rate": 9.683360127014529e-05,
3304
+ "loss": 0.6431,
3305
+ "step": 516
3306
+ },
3307
+ {
3308
+ "epoch": 1.54,
3309
+ "learning_rate": 9.651708366921152e-05,
3310
+ "loss": 0.5867,
3311
+ "step": 517
3312
+ },
3313
+ {
3314
+ "epoch": 1.55,
3315
+ "learning_rate": 9.620060099992609e-05,
3316
+ "loss": 0.574,
3317
+ "step": 518
3318
+ },
3319
+ {
3320
+ "epoch": 1.55,
3321
+ "learning_rate": 9.588415643642826e-05,
3322
+ "loss": 0.5995,
3323
+ "step": 519
3324
+ },
3325
+ {
3326
+ "epoch": 1.55,
3327
+ "learning_rate": 9.556775315247501e-05,
3328
+ "loss": 0.5951,
3329
+ "step": 520
3330
+ },
3331
+ {
3332
+ "epoch": 1.55,
3333
+ "eval_loss": 0.9753234386444092,
3334
+ "eval_runtime": 15.77,
3335
+ "eval_samples_per_second": 36.969,
3336
+ "eval_steps_per_second": 18.516,
3337
+ "step": 520
3338
+ },
3339
+ {
3340
+ "epoch": 1.56,
3341
+ "learning_rate": 9.525139432140945e-05,
3342
+ "loss": 0.5802,
3343
+ "step": 521
3344
+ },
3345
+ {
3346
+ "epoch": 1.56,
3347
+ "learning_rate": 9.493508311612874e-05,
3348
+ "loss": 0.6022,
3349
+ "step": 522
3350
+ },
3351
+ {
3352
+ "epoch": 1.56,
3353
+ "learning_rate": 9.461882270905241e-05,
3354
+ "loss": 0.5621,
3355
+ "step": 523
3356
+ },
3357
+ {
3358
+ "epoch": 1.57,
3359
+ "learning_rate": 9.430261627209044e-05,
3360
+ "loss": 0.5843,
3361
+ "step": 524
3362
+ },
3363
+ {
3364
+ "epoch": 1.57,
3365
+ "learning_rate": 9.398646697661172e-05,
3366
+ "loss": 0.647,
3367
+ "step": 525
3368
+ },
3369
+ {
3370
+ "epoch": 1.57,
3371
+ "learning_rate": 9.367037799341187e-05,
3372
+ "loss": 0.614,
3373
+ "step": 526
3374
+ },
3375
+ {
3376
+ "epoch": 1.57,
3377
+ "learning_rate": 9.335435249268165e-05,
3378
+ "loss": 0.6347,
3379
+ "step": 527
3380
+ },
3381
+ {
3382
+ "epoch": 1.58,
3383
+ "learning_rate": 9.303839364397511e-05,
3384
+ "loss": 0.6657,
3385
+ "step": 528
3386
+ },
3387
+ {
3388
+ "epoch": 1.58,
3389
+ "learning_rate": 9.272250461617793e-05,
3390
+ "loss": 0.6394,
3391
+ "step": 529
3392
+ },
3393
+ {
3394
+ "epoch": 1.58,
3395
+ "learning_rate": 9.24066885774754e-05,
3396
+ "loss": 0.5967,
3397
+ "step": 530
3398
+ },
3399
+ {
3400
+ "epoch": 1.59,
3401
+ "learning_rate": 9.209094869532081e-05,
3402
+ "loss": 0.6504,
3403
+ "step": 531
3404
+ },
3405
+ {
3406
+ "epoch": 1.59,
3407
+ "learning_rate": 9.177528813640362e-05,
3408
+ "loss": 0.5948,
3409
+ "step": 532
3410
+ },
3411
+ {
3412
+ "epoch": 1.59,
3413
+ "learning_rate": 9.145971006661782e-05,
3414
+ "loss": 0.6202,
3415
+ "step": 533
3416
+ },
3417
+ {
3418
+ "epoch": 1.6,
3419
+ "learning_rate": 9.114421765102999e-05,
3420
+ "loss": 0.6011,
3421
+ "step": 534
3422
+ },
3423
+ {
3424
+ "epoch": 1.6,
3425
+ "learning_rate": 9.082881405384765e-05,
3426
+ "loss": 0.6236,
3427
+ "step": 535
3428
+ },
3429
+ {
3430
+ "epoch": 1.6,
3431
+ "learning_rate": 9.051350243838756e-05,
3432
+ "loss": 0.6033,
3433
+ "step": 536
3434
+ },
3435
+ {
3436
+ "epoch": 1.6,
3437
+ "learning_rate": 9.019828596704394e-05,
3438
+ "loss": 0.587,
3439
+ "step": 537
3440
+ },
3441
+ {
3442
+ "epoch": 1.61,
3443
+ "learning_rate": 8.98831678012568e-05,
3444
+ "loss": 0.5835,
3445
+ "step": 538
3446
+ },
3447
+ {
3448
+ "epoch": 1.61,
3449
+ "learning_rate": 8.956815110148012e-05,
3450
+ "loss": 0.5779,
3451
+ "step": 539
3452
+ },
3453
+ {
3454
+ "epoch": 1.61,
3455
+ "learning_rate": 8.925323902715031e-05,
3456
+ "loss": 0.6466,
3457
+ "step": 540
3458
+ },
3459
+ {
3460
+ "epoch": 1.61,
3461
+ "eval_loss": 0.9719182252883911,
3462
+ "eval_runtime": 15.7304,
3463
+ "eval_samples_per_second": 37.062,
3464
+ "eval_steps_per_second": 18.563,
3465
+ "step": 540
3466
+ },
3467
+ {
3468
+ "epoch": 1.62,
3469
+ "learning_rate": 8.893843473665443e-05,
3470
+ "loss": 0.6011,
3471
+ "step": 541
3472
+ },
3473
+ {
3474
+ "epoch": 1.62,
3475
+ "learning_rate": 8.862374138729853e-05,
3476
+ "loss": 0.6446,
3477
+ "step": 542
3478
+ },
3479
+ {
3480
+ "epoch": 1.62,
3481
+ "learning_rate": 8.830916213527592e-05,
3482
+ "loss": 0.6285,
3483
+ "step": 543
3484
+ },
3485
+ {
3486
+ "epoch": 1.63,
3487
+ "learning_rate": 8.799470013563573e-05,
3488
+ "loss": 0.6259,
3489
+ "step": 544
3490
+ },
3491
+ {
3492
+ "epoch": 1.63,
3493
+ "learning_rate": 8.768035854225097e-05,
3494
+ "loss": 0.6143,
3495
+ "step": 545
3496
+ },
3497
+ {
3498
+ "epoch": 1.63,
3499
+ "learning_rate": 8.73661405077871e-05,
3500
+ "loss": 0.6058,
3501
+ "step": 546
3502
+ },
3503
+ {
3504
+ "epoch": 1.63,
3505
+ "learning_rate": 8.705204918367031e-05,
3506
+ "loss": 0.5803,
3507
+ "step": 547
3508
+ },
3509
+ {
3510
+ "epoch": 1.64,
3511
+ "learning_rate": 8.67380877200561e-05,
3512
+ "loss": 0.6259,
3513
+ "step": 548
3514
+ },
3515
+ {
3516
+ "epoch": 1.64,
3517
+ "learning_rate": 8.642425926579739e-05,
3518
+ "loss": 0.5894,
3519
+ "step": 549
3520
+ },
3521
+ {
3522
+ "epoch": 1.64,
3523
+ "learning_rate": 8.611056696841312e-05,
3524
+ "loss": 0.6112,
3525
+ "step": 550
3526
+ },
3527
+ {
3528
+ "epoch": 1.65,
3529
+ "learning_rate": 8.579701397405667e-05,
3530
+ "loss": 0.6618,
3531
+ "step": 551
3532
+ },
3533
+ {
3534
+ "epoch": 1.65,
3535
+ "learning_rate": 8.54836034274844e-05,
3536
+ "loss": 0.6115,
3537
+ "step": 552
3538
+ },
3539
+ {
3540
+ "epoch": 1.65,
3541
+ "learning_rate": 8.517033847202382e-05,
3542
+ "loss": 0.6285,
3543
+ "step": 553
3544
+ },
3545
+ {
3546
+ "epoch": 1.65,
3547
+ "learning_rate": 8.485722224954237e-05,
3548
+ "loss": 0.6507,
3549
+ "step": 554
3550
+ },
3551
+ {
3552
+ "epoch": 1.66,
3553
+ "learning_rate": 8.454425790041567e-05,
3554
+ "loss": 0.6125,
3555
+ "step": 555
3556
+ },
3557
+ {
3558
+ "epoch": 1.66,
3559
+ "learning_rate": 8.423144856349631e-05,
3560
+ "loss": 0.6536,
3561
+ "step": 556
3562
+ },
3563
+ {
3564
+ "epoch": 1.66,
3565
+ "learning_rate": 8.391879737608201e-05,
3566
+ "loss": 0.6476,
3567
+ "step": 557
3568
+ },
3569
+ {
3570
+ "epoch": 1.67,
3571
+ "learning_rate": 8.36063074738845e-05,
3572
+ "loss": 0.6355,
3573
+ "step": 558
3574
+ },
3575
+ {
3576
+ "epoch": 1.67,
3577
+ "learning_rate": 8.329398199099778e-05,
3578
+ "loss": 0.6266,
3579
+ "step": 559
3580
+ },
3581
+ {
3582
+ "epoch": 1.67,
3583
+ "learning_rate": 8.298182405986689e-05,
3584
+ "loss": 0.6246,
3585
+ "step": 560
3586
+ },
3587
+ {
3588
+ "epoch": 1.67,
3589
+ "eval_loss": 0.9681297540664673,
3590
+ "eval_runtime": 15.7562,
3591
+ "eval_samples_per_second": 37.001,
3592
+ "eval_steps_per_second": 18.532,
3593
+ "step": 560
3594
+ },
3595
+ {
3596
+ "epoch": 1.68,
3597
+ "learning_rate": 8.266983681125647e-05,
3598
+ "loss": 0.6251,
3599
+ "step": 561
3600
+ },
3601
+ {
3602
+ "epoch": 1.68,
3603
+ "learning_rate": 8.235802337421919e-05,
3604
+ "loss": 0.6299,
3605
+ "step": 562
3606
+ },
3607
+ {
3608
+ "epoch": 1.68,
3609
+ "learning_rate": 8.20463868760645e-05,
3610
+ "loss": 0.6117,
3611
+ "step": 563
3612
+ },
3613
+ {
3614
+ "epoch": 1.68,
3615
+ "learning_rate": 8.173493044232745e-05,
3616
+ "loss": 0.6027,
3617
+ "step": 564
3618
+ },
3619
+ {
3620
+ "epoch": 1.69,
3621
+ "learning_rate": 8.142365719673693e-05,
3622
+ "loss": 0.6211,
3623
+ "step": 565
3624
+ },
3625
+ {
3626
+ "epoch": 1.69,
3627
+ "learning_rate": 8.11125702611846e-05,
3628
+ "loss": 0.6076,
3629
+ "step": 566
3630
+ },
3631
+ {
3632
+ "epoch": 1.69,
3633
+ "learning_rate": 8.080167275569359e-05,
3634
+ "loss": 0.635,
3635
+ "step": 567
3636
+ },
3637
+ {
3638
+ "epoch": 1.7,
3639
+ "learning_rate": 8.049096779838719e-05,
3640
+ "loss": 0.603,
3641
+ "step": 568
3642
+ },
3643
+ {
3644
+ "epoch": 1.7,
3645
+ "learning_rate": 8.018045850545743e-05,
3646
+ "loss": 0.6406,
3647
+ "step": 569
3648
+ },
3649
+ {
3650
+ "epoch": 1.7,
3651
+ "learning_rate": 7.987014799113397e-05,
3652
+ "loss": 0.5941,
3653
+ "step": 570
3654
+ },
3655
+ {
3656
+ "epoch": 1.71,
3657
+ "learning_rate": 7.956003936765297e-05,
3658
+ "loss": 0.5542,
3659
+ "step": 571
3660
+ },
3661
+ {
3662
+ "epoch": 1.71,
3663
+ "learning_rate": 7.925013574522557e-05,
3664
+ "loss": 0.5938,
3665
+ "step": 572
3666
+ },
3667
+ {
3668
+ "epoch": 1.71,
3669
+ "learning_rate": 7.894044023200691e-05,
3670
+ "loss": 0.5701,
3671
+ "step": 573
3672
+ },
3673
+ {
3674
+ "epoch": 1.71,
3675
+ "learning_rate": 7.863095593406491e-05,
3676
+ "loss": 0.6276,
3677
+ "step": 574
3678
+ },
3679
+ {
3680
+ "epoch": 1.72,
3681
+ "learning_rate": 7.832168595534924e-05,
3682
+ "loss": 0.6352,
3683
+ "step": 575
3684
+ },
3685
+ {
3686
+ "epoch": 1.72,
3687
+ "learning_rate": 7.801263339765994e-05,
3688
+ "loss": 0.5628,
3689
+ "step": 576
3690
+ },
3691
+ {
3692
+ "epoch": 1.72,
3693
+ "learning_rate": 7.770380136061643e-05,
3694
+ "loss": 0.6285,
3695
+ "step": 577
3696
+ },
3697
+ {
3698
+ "epoch": 1.73,
3699
+ "learning_rate": 7.739519294162652e-05,
3700
+ "loss": 0.6409,
3701
+ "step": 578
3702
+ },
3703
+ {
3704
+ "epoch": 1.73,
3705
+ "learning_rate": 7.708681123585525e-05,
3706
+ "loss": 0.6944,
3707
+ "step": 579
3708
+ },
3709
+ {
3710
+ "epoch": 1.73,
3711
+ "learning_rate": 7.677865933619379e-05,
3712
+ "loss": 0.627,
3713
+ "step": 580
3714
+ },
3715
+ {
3716
+ "epoch": 1.73,
3717
+ "eval_loss": 0.9704866409301758,
3718
+ "eval_runtime": 15.7519,
3719
+ "eval_samples_per_second": 37.011,
3720
+ "eval_steps_per_second": 18.537,
3721
+ "step": 580
3722
+ },
3723
+ {
3724
+ "epoch": 1.74,
3725
+ "learning_rate": 7.647074033322856e-05,
3726
+ "loss": 0.656,
3727
+ "step": 581
3728
+ },
3729
+ {
3730
+ "epoch": 1.74,
3731
+ "learning_rate": 7.616305731521008e-05,
3732
+ "loss": 0.6058,
3733
+ "step": 582
3734
+ },
3735
+ {
3736
+ "epoch": 1.74,
3737
+ "learning_rate": 7.585561336802221e-05,
3738
+ "loss": 0.5871,
3739
+ "step": 583
3740
+ },
3741
+ {
3742
+ "epoch": 1.74,
3743
+ "learning_rate": 7.554841157515092e-05,
3744
+ "loss": 0.6347,
3745
+ "step": 584
3746
+ },
3747
+ {
3748
+ "epoch": 1.75,
3749
+ "learning_rate": 7.524145501765357e-05,
3750
+ "loss": 0.5847,
3751
+ "step": 585
3752
+ },
3753
+ {
3754
+ "epoch": 1.75,
3755
+ "learning_rate": 7.493474677412794e-05,
3756
+ "loss": 0.5988,
3757
+ "step": 586
3758
+ },
3759
+ {
3760
+ "epoch": 1.75,
3761
+ "learning_rate": 7.462828992068143e-05,
3762
+ "loss": 0.5942,
3763
+ "step": 587
3764
+ },
3765
+ {
3766
+ "epoch": 1.76,
3767
+ "learning_rate": 7.432208753090009e-05,
3768
+ "loss": 0.6388,
3769
+ "step": 588
3770
+ },
3771
+ {
3772
+ "epoch": 1.76,
3773
+ "learning_rate": 7.401614267581779e-05,
3774
+ "loss": 0.6184,
3775
+ "step": 589
3776
+ },
3777
+ {
3778
+ "epoch": 1.76,
3779
+ "learning_rate": 7.371045842388552e-05,
3780
+ "loss": 0.5469,
3781
+ "step": 590
3782
+ },
3783
+ {
3784
+ "epoch": 1.77,
3785
+ "learning_rate": 7.340503784094066e-05,
3786
+ "loss": 0.591,
3787
+ "step": 591
3788
+ },
3789
+ {
3790
+ "epoch": 1.77,
3791
+ "learning_rate": 7.309988399017602e-05,
3792
+ "loss": 0.64,
3793
+ "step": 592
3794
+ },
3795
+ {
3796
+ "epoch": 1.77,
3797
+ "learning_rate": 7.279499993210928e-05,
3798
+ "loss": 0.6149,
3799
+ "step": 593
3800
+ },
3801
+ {
3802
+ "epoch": 1.77,
3803
+ "learning_rate": 7.24903887245522e-05,
3804
+ "loss": 0.7015,
3805
+ "step": 594
3806
+ },
3807
+ {
3808
+ "epoch": 1.78,
3809
+ "learning_rate": 7.218605342258015e-05,
3810
+ "loss": 0.6292,
3811
+ "step": 595
3812
+ },
3813
+ {
3814
+ "epoch": 1.78,
3815
+ "learning_rate": 7.188199707850122e-05,
3816
+ "loss": 0.6177,
3817
+ "step": 596
3818
+ },
3819
+ {
3820
+ "epoch": 1.78,
3821
+ "learning_rate": 7.15782227418257e-05,
3822
+ "loss": 0.5794,
3823
+ "step": 597
3824
+ },
3825
+ {
3826
+ "epoch": 1.79,
3827
+ "learning_rate": 7.127473345923554e-05,
3828
+ "loss": 0.6237,
3829
+ "step": 598
3830
+ },
3831
+ {
3832
+ "epoch": 1.79,
3833
+ "learning_rate": 7.097153227455379e-05,
3834
+ "loss": 0.6117,
3835
+ "step": 599
3836
+ },
3837
+ {
3838
+ "epoch": 1.79,
3839
+ "learning_rate": 7.066862222871397e-05,
3840
+ "loss": 0.6214,
3841
+ "step": 600
3842
+ },
3843
+ {
3844
+ "epoch": 1.79,
3845
+ "eval_loss": 0.9691401720046997,
3846
+ "eval_runtime": 15.7684,
3847
+ "eval_samples_per_second": 36.973,
3848
+ "eval_steps_per_second": 18.518,
3849
+ "step": 600
3850
+ },
3851
+ {
3852
+ "epoch": 1.8,
3853
+ "learning_rate": 7.03660063597297e-05,
3854
+ "loss": 0.5839,
3855
+ "step": 601
3856
+ },
3857
+ {
3858
+ "epoch": 1.8,
3859
+ "learning_rate": 7.006368770266421e-05,
3860
+ "loss": 0.5862,
3861
+ "step": 602
3862
+ },
3863
+ {
3864
+ "epoch": 1.8,
3865
+ "learning_rate": 6.976166928959977e-05,
3866
+ "loss": 0.5901,
3867
+ "step": 603
3868
+ },
3869
+ {
3870
+ "epoch": 1.8,
3871
+ "learning_rate": 6.945995414960744e-05,
3872
+ "loss": 0.6055,
3873
+ "step": 604
3874
+ },
3875
+ {
3876
+ "epoch": 1.81,
3877
+ "learning_rate": 6.915854530871658e-05,
3878
+ "loss": 0.5828,
3879
+ "step": 605
3880
+ },
3881
+ {
3882
+ "epoch": 1.81,
3883
+ "learning_rate": 6.885744578988463e-05,
3884
+ "loss": 0.5737,
3885
+ "step": 606
3886
+ },
3887
+ {
3888
+ "epoch": 1.81,
3889
+ "learning_rate": 6.855665861296661e-05,
3890
+ "loss": 0.5896,
3891
+ "step": 607
3892
+ },
3893
+ {
3894
+ "epoch": 1.82,
3895
+ "learning_rate": 6.825618679468502e-05,
3896
+ "loss": 0.5914,
3897
+ "step": 608
3898
+ },
3899
+ {
3900
+ "epoch": 1.82,
3901
+ "learning_rate": 6.795603334859935e-05,
3902
+ "loss": 0.6071,
3903
+ "step": 609
3904
+ },
3905
+ {
3906
+ "epoch": 1.82,
3907
+ "learning_rate": 6.765620128507619e-05,
3908
+ "loss": 0.6257,
3909
+ "step": 610
3910
+ },
3911
+ {
3912
+ "epoch": 1.83,
3913
+ "learning_rate": 6.735669361125873e-05,
3914
+ "loss": 0.6182,
3915
+ "step": 611
3916
+ },
3917
+ {
3918
+ "epoch": 1.83,
3919
+ "learning_rate": 6.705751333103675e-05,
3920
+ "loss": 0.5525,
3921
+ "step": 612
3922
+ },
3923
+ {
3924
+ "epoch": 1.83,
3925
+ "learning_rate": 6.67586634450164e-05,
3926
+ "loss": 0.5787,
3927
+ "step": 613
3928
+ },
3929
+ {
3930
+ "epoch": 1.83,
3931
+ "learning_rate": 6.64601469504903e-05,
3932
+ "loss": 0.5781,
3933
+ "step": 614
3934
+ },
3935
+ {
3936
+ "epoch": 1.84,
3937
+ "learning_rate": 6.61619668414072e-05,
3938
+ "loss": 0.5784,
3939
+ "step": 615
3940
+ },
3941
+ {
3942
+ "epoch": 1.84,
3943
+ "learning_rate": 6.586412610834221e-05,
3944
+ "loss": 0.6093,
3945
+ "step": 616
3946
+ },
3947
+ {
3948
+ "epoch": 1.84,
3949
+ "learning_rate": 6.556662773846658e-05,
3950
+ "loss": 0.5732,
3951
+ "step": 617
3952
+ },
3953
+ {
3954
+ "epoch": 1.85,
3955
+ "learning_rate": 6.526947471551798e-05,
3956
+ "loss": 0.6163,
3957
+ "step": 618
3958
+ },
3959
+ {
3960
+ "epoch": 1.85,
3961
+ "learning_rate": 6.497267001977037e-05,
3962
+ "loss": 0.5984,
3963
+ "step": 619
3964
+ },
3965
+ {
3966
+ "epoch": 1.85,
3967
+ "learning_rate": 6.46762166280042e-05,
3968
+ "loss": 0.6558,
3969
+ "step": 620
3970
+ },
3971
+ {
3972
+ "epoch": 1.85,
3973
+ "eval_loss": 0.9709066152572632,
3974
+ "eval_runtime": 15.7265,
3975
+ "eval_samples_per_second": 37.071,
3976
+ "eval_steps_per_second": 18.567,
3977
+ "step": 620
3978
+ },
3979
+ {
3980
+ "epoch": 1.86,
3981
+ "learning_rate": 6.438011751347655e-05,
3982
+ "loss": 0.596,
3983
+ "step": 621
3984
+ },
3985
+ {
3986
+ "epoch": 1.86,
3987
+ "learning_rate": 6.40843756458913e-05,
3988
+ "loss": 0.5747,
3989
+ "step": 622
3990
+ },
3991
+ {
3992
+ "epoch": 1.86,
3993
+ "learning_rate": 6.378899399136938e-05,
3994
+ "loss": 0.6224,
3995
+ "step": 623
3996
+ },
3997
+ {
3998
+ "epoch": 1.86,
3999
+ "learning_rate": 6.349397551241894e-05,
4000
+ "loss": 0.6119,
4001
+ "step": 624
4002
+ },
4003
+ {
4004
+ "epoch": 1.87,
4005
+ "learning_rate": 6.319932316790565e-05,
4006
+ "loss": 0.632,
4007
+ "step": 625
4008
+ },
4009
+ {
4010
+ "epoch": 1.87,
4011
+ "learning_rate": 6.290503991302324e-05,
4012
+ "loss": 0.6042,
4013
+ "step": 626
4014
+ },
4015
+ {
4016
+ "epoch": 1.87,
4017
+ "learning_rate": 6.261112869926347e-05,
4018
+ "loss": 0.5992,
4019
+ "step": 627
4020
+ },
4021
+ {
4022
+ "epoch": 1.88,
4023
+ "learning_rate": 6.231759247438689e-05,
4024
+ "loss": 0.6207,
4025
+ "step": 628
4026
+ },
4027
+ {
4028
+ "epoch": 1.88,
4029
+ "learning_rate": 6.202443418239302e-05,
4030
+ "loss": 0.6841,
4031
+ "step": 629
4032
+ },
4033
+ {
4034
+ "epoch": 1.88,
4035
+ "learning_rate": 6.173165676349103e-05,
4036
+ "loss": 0.616,
4037
+ "step": 630
4038
+ },
4039
+ {
4040
+ "epoch": 1.88,
4041
+ "learning_rate": 6.14392631540701e-05,
4042
+ "loss": 0.61,
4043
+ "step": 631
4044
+ },
4045
+ {
4046
+ "epoch": 1.89,
4047
+ "learning_rate": 6.114725628666998e-05,
4048
+ "loss": 0.587,
4049
+ "step": 632
4050
+ },
4051
+ {
4052
+ "epoch": 1.89,
4053
+ "learning_rate": 6.0855639089951735e-05,
4054
+ "loss": 0.6161,
4055
+ "step": 633
4056
+ },
4057
+ {
4058
+ "epoch": 1.89,
4059
+ "learning_rate": 6.0564414488668165e-05,
4060
+ "loss": 0.6047,
4061
+ "step": 634
4062
+ },
4063
+ {
4064
+ "epoch": 1.9,
4065
+ "learning_rate": 6.027358540363457e-05,
4066
+ "loss": 0.6326,
4067
+ "step": 635
4068
+ },
4069
+ {
4070
+ "epoch": 1.9,
4071
+ "learning_rate": 5.998315475169942e-05,
4072
+ "loss": 0.6319,
4073
+ "step": 636
4074
+ },
4075
+ {
4076
+ "epoch": 1.9,
4077
+ "learning_rate": 5.969312544571529e-05,
4078
+ "loss": 0.5422,
4079
+ "step": 637
4080
+ },
4081
+ {
4082
+ "epoch": 1.91,
4083
+ "learning_rate": 5.94035003945093e-05,
4084
+ "loss": 0.5769,
4085
+ "step": 638
4086
+ },
4087
+ {
4088
+ "epoch": 1.91,
4089
+ "learning_rate": 5.911428250285421e-05,
4090
+ "loss": 0.6539,
4091
+ "step": 639
4092
+ },
4093
+ {
4094
+ "epoch": 1.91,
4095
+ "learning_rate": 5.88254746714392e-05,
4096
+ "loss": 0.5736,
4097
+ "step": 640
4098
+ },
4099
+ {
4100
+ "epoch": 1.91,
4101
+ "eval_loss": 0.9673944711685181,
4102
+ "eval_runtime": 22.4205,
4103
+ "eval_samples_per_second": 26.003,
4104
+ "eval_steps_per_second": 13.024,
4105
+ "step": 640
4106
+ },
4107
+ {
4108
+ "epoch": 1.91,
4109
+ "learning_rate": 5.8537079796840765e-05,
4110
+ "loss": 0.6077,
4111
+ "step": 641
4112
+ },
4113
+ {
4114
+ "epoch": 1.92,
4115
+ "learning_rate": 5.824910077149371e-05,
4116
+ "loss": 0.6218,
4117
+ "step": 642
4118
+ },
4119
+ {
4120
+ "epoch": 1.92,
4121
+ "learning_rate": 5.796154048366208e-05,
4122
+ "loss": 0.6631,
4123
+ "step": 643
4124
+ },
4125
+ {
4126
+ "epoch": 1.92,
4127
+ "learning_rate": 5.767440181741019e-05,
4128
+ "loss": 0.596,
4129
+ "step": 644
4130
+ },
4131
+ {
4132
+ "epoch": 1.93,
4133
+ "learning_rate": 5.7387687652573865e-05,
4134
+ "loss": 0.6009,
4135
+ "step": 645
4136
+ },
4137
+ {
4138
+ "epoch": 1.93,
4139
+ "learning_rate": 5.710140086473129e-05,
4140
+ "loss": 0.6687,
4141
+ "step": 646
4142
+ },
4143
+ {
4144
+ "epoch": 1.93,
4145
+ "learning_rate": 5.6815544325174355e-05,
4146
+ "loss": 0.6074,
4147
+ "step": 647
4148
+ },
4149
+ {
4150
+ "epoch": 1.94,
4151
+ "learning_rate": 5.653012090087977e-05,
4152
+ "loss": 0.6322,
4153
+ "step": 648
4154
+ },
4155
+ {
4156
+ "epoch": 1.94,
4157
+ "learning_rate": 5.6245133454480446e-05,
4158
+ "loss": 0.6062,
4159
+ "step": 649
4160
+ },
4161
+ {
4162
+ "epoch": 1.94,
4163
+ "learning_rate": 5.596058484423656e-05,
4164
+ "loss": 0.6011,
4165
+ "step": 650
4166
+ },
4167
+ {
4168
+ "epoch": 1.94,
4169
+ "learning_rate": 5.5676477924007166e-05,
4170
+ "loss": 0.6329,
4171
+ "step": 651
4172
+ },
4173
+ {
4174
+ "epoch": 1.95,
4175
+ "learning_rate": 5.5392815543221254e-05,
4176
+ "loss": 0.6324,
4177
+ "step": 652
4178
+ },
4179
+ {
4180
+ "epoch": 1.95,
4181
+ "learning_rate": 5.5109600546849416e-05,
4182
+ "loss": 0.638,
4183
+ "step": 653
4184
+ },
4185
+ {
4186
+ "epoch": 1.95,
4187
+ "learning_rate": 5.4826835775375285e-05,
4188
+ "loss": 0.6016,
4189
+ "step": 654
4190
+ },
4191
+ {
4192
+ "epoch": 1.96,
4193
+ "learning_rate": 5.454452406476694e-05,
4194
+ "loss": 0.6221,
4195
+ "step": 655
4196
+ },
4197
+ {
4198
+ "epoch": 1.96,
4199
+ "learning_rate": 5.4262668246448475e-05,
4200
+ "loss": 0.6496,
4201
+ "step": 656
4202
+ },
4203
+ {
4204
+ "epoch": 1.96,
4205
+ "learning_rate": 5.398127114727178e-05,
4206
+ "loss": 0.6118,
4207
+ "step": 657
4208
+ },
4209
+ {
4210
+ "epoch": 1.97,
4211
+ "learning_rate": 5.3700335589487925e-05,
4212
+ "loss": 0.6216,
4213
+ "step": 658
4214
+ },
4215
+ {
4216
+ "epoch": 1.97,
4217
+ "learning_rate": 5.341986439071907e-05,
4218
+ "loss": 0.6238,
4219
+ "step": 659
4220
+ },
4221
+ {
4222
+ "epoch": 1.97,
4223
+ "learning_rate": 5.3139860363929996e-05,
4224
+ "loss": 0.6188,
4225
+ "step": 660
4226
+ },
4227
+ {
4228
+ "epoch": 1.97,
4229
+ "eval_loss": 0.9674209356307983,
4230
+ "eval_runtime": 15.74,
4231
+ "eval_samples_per_second": 37.039,
4232
+ "eval_steps_per_second": 18.552,
4233
+ "step": 660
4234
+ },
4235
+ {
4236
+ "epoch": 1.97,
4237
+ "learning_rate": 5.286032631740023e-05,
4238
+ "loss": 0.6111,
4239
+ "step": 661
4240
+ },
4241
+ {
4242
+ "epoch": 1.98,
4243
+ "learning_rate": 5.2581265054695494e-05,
4244
+ "loss": 0.6158,
4245
+ "step": 662
4246
+ },
4247
+ {
4248
+ "epoch": 1.98,
4249
+ "learning_rate": 5.230267937463977e-05,
4250
+ "loss": 0.6135,
4251
+ "step": 663
4252
+ },
4253
+ {
4254
+ "epoch": 1.98,
4255
+ "learning_rate": 5.202457207128736e-05,
4256
+ "loss": 0.5729,
4257
+ "step": 664
4258
+ },
4259
+ {
4260
+ "epoch": 1.99,
4261
+ "learning_rate": 5.174694593389455e-05,
4262
+ "loss": 0.5397,
4263
+ "step": 665
4264
+ },
4265
+ {
4266
+ "epoch": 1.99,
4267
+ "learning_rate": 5.146980374689192e-05,
4268
+ "loss": 0.6445,
4269
+ "step": 666
4270
+ },
4271
+ {
4272
+ "epoch": 1.99,
4273
+ "learning_rate": 5.119314828985619e-05,
4274
+ "loss": 0.6251,
4275
+ "step": 667
4276
+ },
4277
+ {
4278
+ "epoch": 2.0,
4279
+ "learning_rate": 5.0916982337482644e-05,
4280
+ "loss": 0.5926,
4281
+ "step": 668
4282
+ },
4283
+ {
4284
+ "epoch": 2.0,
4285
+ "learning_rate": 5.0641308659556964e-05,
4286
+ "loss": 0.6072,
4287
+ "step": 669
4288
+ }
4289
+ ],
4290
+ "logging_steps": 1,
4291
+ "max_steps": 1002,
4292
+ "num_train_epochs": 3,
4293
+ "save_steps": 500,
4294
+ "total_flos": 9.470113214952899e+17,
4295
+ "trial_name": null,
4296
+ "trial_params": null
4297
+ }
checkpoint-669/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7780122e038bb83236a6b213b013867a89c4e742cce8c1208c1a41fc54e6c57
3
+ size 4475
config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "mistralai/Mistral-7B-v0.1",
3
+ "architectures": [
4
+ "MistralForCausalLM"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 4096,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 14336,
12
+ "max_position_embeddings": 32768,
13
+ "model_type": "mistral",
14
+ "num_attention_heads": 32,
15
+ "num_hidden_layers": 32,
16
+ "num_key_value_heads": 8,
17
+ "quantization_config": {
18
+ "bnb_4bit_compute_dtype": "float32",
19
+ "bnb_4bit_quant_type": "fp4",
20
+ "bnb_4bit_use_double_quant": false,
21
+ "llm_int8_enable_fp32_cpu_offload": false,
22
+ "llm_int8_has_fp16_weight": false,
23
+ "llm_int8_skip_modules": null,
24
+ "llm_int8_threshold": 6.0,
25
+ "load_in_4bit": false,
26
+ "load_in_8bit": true,
27
+ "quant_method": "bitsandbytes"
28
+ },
29
+ "rms_norm_eps": 1e-05,
30
+ "rope_theta": 10000.0,
31
+ "sliding_window": 4096,
32
+ "tie_word_embeddings": false,
33
+ "torch_dtype": "bfloat16",
34
+ "transformers_version": "4.34.0.dev0",
35
+ "use_cache": false,
36
+ "vocab_size": 32000
37
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "pad_token": "</s>",
5
+ "unk_token": "<unk>"
6
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": true,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": true,
16
+ "normalized": false,
17
+ "rstrip": true,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": true,
24
+ "normalized": false,
25
+ "rstrip": true,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": true,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": null,
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "tokenizer_file": "/root/.cache/huggingface/hub/models--mistralai--Mistral-7B-v0.1/snapshots/78814a934d8ce59ee93599378f3e929ff137da06/tokenizer.json",
41
+ "trust_remote_code": false,
42
+ "unk_token": "<unk>",
43
+ "use_default_system_prompt": true,
44
+ "use_fast": true
45
+ }