Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep22_20-43-20_COE-CS-sv003/events.out.tfevents.1727038103.COE-CS-sv003.713618.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2667 -0
README.md
ADDED
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
5 |
+
tags:
|
6 |
+
- trl
|
7 |
+
- sft
|
8 |
+
- generated_from_trainer
|
9 |
+
model-index:
|
10 |
+
- name: feedback_p0.1_seed42_level2_rarecleanbatch16_reverse
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# feedback_p0.1_seed42_level2_rarecleanbatch16_reverse
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the None dataset.
|
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: 1e-05
|
39 |
+
- train_batch_size: 4
|
40 |
+
- eval_batch_size: 1
|
41 |
+
- seed: 42
|
42 |
+
- distributed_type: multi-GPU
|
43 |
+
- num_devices: 4
|
44 |
+
- total_train_batch_size: 16
|
45 |
+
- total_eval_batch_size: 4
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: cosine
|
48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
49 |
+
- num_epochs: 1
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
|
54 |
+
|
55 |
+
### Framework versions
|
56 |
+
|
57 |
+
- Transformers 4.44.2
|
58 |
+
- Pytorch 2.3.1+cu121
|
59 |
+
- Datasets 2.19.1
|
60 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 195665288232960.0,
|
4 |
+
"train_loss": 0.4974420365322842,
|
5 |
+
"train_runtime": 16839.356,
|
6 |
+
"train_samples": 98952,
|
7 |
+
"train_samples_per_second": 1.776,
|
8 |
+
"train_steps_per_second": 0.111
|
9 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"transformers_version": "4.44.2"
|
6 |
+
}
|
runs/Sep22_20-43-20_COE-CS-sv003/events.out.tfevents.1727038103.COE-CS-sv003.713618.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:40076737b717f44b4f48efb5b1a52a7f34611a80b3f85d0c44ea335c081a1ee9
|
3 |
+
size 84685
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 195665288232960.0,
|
4 |
+
"train_loss": 0.4974420365322842,
|
5 |
+
"train_runtime": 16839.356,
|
6 |
+
"train_samples": 98952,
|
7 |
+
"train_samples_per_second": 1.776,
|
8 |
+
"train_steps_per_second": 0.111
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2667 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1869,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0005350454788657035,
|
13 |
+
"grad_norm": 23.32975499538448,
|
14 |
+
"learning_rate": 5.3475935828877005e-08,
|
15 |
+
"loss": 1.3233,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.002675227394328518,
|
20 |
+
"grad_norm": 22.00335872083755,
|
21 |
+
"learning_rate": 2.6737967914438503e-07,
|
22 |
+
"loss": 1.3186,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.005350454788657036,
|
27 |
+
"grad_norm": 15.4065659696612,
|
28 |
+
"learning_rate": 5.347593582887701e-07,
|
29 |
+
"loss": 1.2822,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.008025682182985553,
|
34 |
+
"grad_norm": 11.502912896830692,
|
35 |
+
"learning_rate": 8.021390374331551e-07,
|
36 |
+
"loss": 1.1274,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.010700909577314071,
|
41 |
+
"grad_norm": 9.500638744694992,
|
42 |
+
"learning_rate": 1.0695187165775401e-06,
|
43 |
+
"loss": 1.0332,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.01337613697164259,
|
48 |
+
"grad_norm": 3.5746284262621084,
|
49 |
+
"learning_rate": 1.3368983957219254e-06,
|
50 |
+
"loss": 0.908,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.016051364365971106,
|
55 |
+
"grad_norm": 3.476299775893293,
|
56 |
+
"learning_rate": 1.6042780748663103e-06,
|
57 |
+
"loss": 0.8905,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.018726591760299626,
|
62 |
+
"grad_norm": 2.9586982678137983,
|
63 |
+
"learning_rate": 1.8716577540106954e-06,
|
64 |
+
"loss": 0.8566,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.021401819154628143,
|
69 |
+
"grad_norm": 2.921422317023882,
|
70 |
+
"learning_rate": 2.1390374331550802e-06,
|
71 |
+
"loss": 0.8463,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.024077046548956663,
|
76 |
+
"grad_norm": 2.8850607045290033,
|
77 |
+
"learning_rate": 2.4064171122994653e-06,
|
78 |
+
"loss": 0.8306,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.02675227394328518,
|
83 |
+
"grad_norm": 3.204350090952928,
|
84 |
+
"learning_rate": 2.673796791443851e-06,
|
85 |
+
"loss": 0.813,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.029427501337613696,
|
90 |
+
"grad_norm": 3.082541545860967,
|
91 |
+
"learning_rate": 2.9411764705882355e-06,
|
92 |
+
"loss": 0.8103,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.03210272873194221,
|
97 |
+
"grad_norm": 3.03497970770737,
|
98 |
+
"learning_rate": 3.2085561497326205e-06,
|
99 |
+
"loss": 0.7921,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.034777956126270736,
|
104 |
+
"grad_norm": 3.102342967951045,
|
105 |
+
"learning_rate": 3.4759358288770056e-06,
|
106 |
+
"loss": 0.7957,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.03745318352059925,
|
111 |
+
"grad_norm": 2.9578938537372044,
|
112 |
+
"learning_rate": 3.7433155080213907e-06,
|
113 |
+
"loss": 0.7839,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.04012841091492777,
|
118 |
+
"grad_norm": 3.0373015894615283,
|
119 |
+
"learning_rate": 4.010695187165775e-06,
|
120 |
+
"loss": 0.773,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.042803638309256285,
|
125 |
+
"grad_norm": 3.076413211099657,
|
126 |
+
"learning_rate": 4.2780748663101604e-06,
|
127 |
+
"loss": 0.7879,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.0454788657035848,
|
132 |
+
"grad_norm": 3.116251894957883,
|
133 |
+
"learning_rate": 4.5454545454545455e-06,
|
134 |
+
"loss": 0.7564,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.048154093097913325,
|
139 |
+
"grad_norm": 3.293157475867486,
|
140 |
+
"learning_rate": 4.812834224598931e-06,
|
141 |
+
"loss": 0.7447,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.05082932049224184,
|
146 |
+
"grad_norm": 2.9781238058172015,
|
147 |
+
"learning_rate": 5.0802139037433165e-06,
|
148 |
+
"loss": 0.7333,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.05350454788657036,
|
153 |
+
"grad_norm": 2.954596587676888,
|
154 |
+
"learning_rate": 5.347593582887702e-06,
|
155 |
+
"loss": 0.7373,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.056179775280898875,
|
160 |
+
"grad_norm": 2.8773857530480833,
|
161 |
+
"learning_rate": 5.614973262032086e-06,
|
162 |
+
"loss": 0.7303,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.05885500267522739,
|
167 |
+
"grad_norm": 2.9567813017811084,
|
168 |
+
"learning_rate": 5.882352941176471e-06,
|
169 |
+
"loss": 0.7279,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.061530230069555915,
|
174 |
+
"grad_norm": 2.9858797822378955,
|
175 |
+
"learning_rate": 6.149732620320856e-06,
|
176 |
+
"loss": 0.7214,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.06420545746388442,
|
181 |
+
"grad_norm": 3.012701625430249,
|
182 |
+
"learning_rate": 6.417112299465241e-06,
|
183 |
+
"loss": 0.717,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.06688068485821295,
|
188 |
+
"grad_norm": 2.9708435052571955,
|
189 |
+
"learning_rate": 6.684491978609626e-06,
|
190 |
+
"loss": 0.7248,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.06955591225254147,
|
195 |
+
"grad_norm": 2.8861878079531236,
|
196 |
+
"learning_rate": 6.951871657754011e-06,
|
197 |
+
"loss": 0.72,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.07223113964686999,
|
202 |
+
"grad_norm": 2.887798712701533,
|
203 |
+
"learning_rate": 7.219251336898396e-06,
|
204 |
+
"loss": 0.7172,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.0749063670411985,
|
209 |
+
"grad_norm": 2.9349544237124094,
|
210 |
+
"learning_rate": 7.486631016042781e-06,
|
211 |
+
"loss": 0.711,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.07758159443552702,
|
216 |
+
"grad_norm": 3.0198425274374943,
|
217 |
+
"learning_rate": 7.754010695187166e-06,
|
218 |
+
"loss": 0.7007,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.08025682182985554,
|
223 |
+
"grad_norm": 2.844285698601967,
|
224 |
+
"learning_rate": 8.02139037433155e-06,
|
225 |
+
"loss": 0.7163,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.08293204922418405,
|
230 |
+
"grad_norm": 2.906658583579405,
|
231 |
+
"learning_rate": 8.288770053475937e-06,
|
232 |
+
"loss": 0.7135,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.08560727661851257,
|
237 |
+
"grad_norm": 2.775520126426031,
|
238 |
+
"learning_rate": 8.556149732620321e-06,
|
239 |
+
"loss": 0.6986,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.08828250401284109,
|
244 |
+
"grad_norm": 2.69467913452044,
|
245 |
+
"learning_rate": 8.823529411764707e-06,
|
246 |
+
"loss": 0.6965,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.0909577314071696,
|
251 |
+
"grad_norm": 2.7282250830391357,
|
252 |
+
"learning_rate": 9.090909090909091e-06,
|
253 |
+
"loss": 0.7034,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.09363295880149813,
|
258 |
+
"grad_norm": 2.892650307618601,
|
259 |
+
"learning_rate": 9.358288770053477e-06,
|
260 |
+
"loss": 0.7118,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.09630818619582665,
|
265 |
+
"grad_norm": 2.6740856111003164,
|
266 |
+
"learning_rate": 9.625668449197861e-06,
|
267 |
+
"loss": 0.6921,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.09898341359015517,
|
272 |
+
"grad_norm": 2.833027188962373,
|
273 |
+
"learning_rate": 9.893048128342247e-06,
|
274 |
+
"loss": 0.7014,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.10165864098448368,
|
279 |
+
"grad_norm": 3.0007083274918993,
|
280 |
+
"learning_rate": 9.999921507322408e-06,
|
281 |
+
"loss": 0.6943,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.1043338683788122,
|
286 |
+
"grad_norm": 2.699937414637709,
|
287 |
+
"learning_rate": 9.999441838772916e-06,
|
288 |
+
"loss": 0.6879,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.10700909577314072,
|
293 |
+
"grad_norm": 2.8840558663751246,
|
294 |
+
"learning_rate": 9.99852615049999e-06,
|
295 |
+
"loss": 0.6934,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.10968432316746923,
|
300 |
+
"grad_norm": 2.6349082313416026,
|
301 |
+
"learning_rate": 9.997174522364177e-06,
|
302 |
+
"loss": 0.6857,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.11235955056179775,
|
307 |
+
"grad_norm": 2.7831287150153137,
|
308 |
+
"learning_rate": 9.995387072245939e-06,
|
309 |
+
"loss": 0.7051,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.11503477795612627,
|
314 |
+
"grad_norm": 2.869759215432631,
|
315 |
+
"learning_rate": 9.993163956035381e-06,
|
316 |
+
"loss": 0.698,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.11771000535045478,
|
321 |
+
"grad_norm": 2.6687295223771885,
|
322 |
+
"learning_rate": 9.990505367618647e-06,
|
323 |
+
"loss": 0.6688,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.12038523274478331,
|
328 |
+
"grad_norm": 2.8932052543699442,
|
329 |
+
"learning_rate": 9.987411538861023e-06,
|
330 |
+
"loss": 0.6748,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.12306046013911183,
|
335 |
+
"grad_norm": 2.6718609572340806,
|
336 |
+
"learning_rate": 9.9838827395867e-06,
|
337 |
+
"loss": 0.6905,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.12573568753344033,
|
342 |
+
"grad_norm": 2.812043760617457,
|
343 |
+
"learning_rate": 9.979919277555247e-06,
|
344 |
+
"loss": 0.6768,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.12841091492776885,
|
349 |
+
"grad_norm": 3.211381659481926,
|
350 |
+
"learning_rate": 9.97552149843478e-06,
|
351 |
+
"loss": 0.6787,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.13108614232209737,
|
356 |
+
"grad_norm": 3.1485034799180593,
|
357 |
+
"learning_rate": 9.970689785771798e-06,
|
358 |
+
"loss": 0.6815,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.1337613697164259,
|
363 |
+
"grad_norm": 2.7247090563062417,
|
364 |
+
"learning_rate": 9.96542456095775e-06,
|
365 |
+
"loss": 0.6819,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.13643659711075443,
|
370 |
+
"grad_norm": 2.7085417921956387,
|
371 |
+
"learning_rate": 9.95972628319227e-06,
|
372 |
+
"loss": 0.6681,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.13911182450508294,
|
377 |
+
"grad_norm": 2.691457151596662,
|
378 |
+
"learning_rate": 9.953595449443134e-06,
|
379 |
+
"loss": 0.6579,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.14178705189941146,
|
384 |
+
"grad_norm": 2.584599998752218,
|
385 |
+
"learning_rate": 9.947032594402917e-06,
|
386 |
+
"loss": 0.679,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.14446227929373998,
|
391 |
+
"grad_norm": 2.6045453255332554,
|
392 |
+
"learning_rate": 9.940038290442362e-06,
|
393 |
+
"loss": 0.6556,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.1471375066880685,
|
398 |
+
"grad_norm": 2.4412081165922848,
|
399 |
+
"learning_rate": 9.932613147560464e-06,
|
400 |
+
"loss": 0.6703,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.149812734082397,
|
405 |
+
"grad_norm": 2.5230097371528437,
|
406 |
+
"learning_rate": 9.924757813331256e-06,
|
407 |
+
"loss": 0.6744,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.15248796147672553,
|
412 |
+
"grad_norm": 2.585452773043538,
|
413 |
+
"learning_rate": 9.916472972847353e-06,
|
414 |
+
"loss": 0.669,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.15516318887105404,
|
419 |
+
"grad_norm": 2.645498442811453,
|
420 |
+
"learning_rate": 9.907759348660186e-06,
|
421 |
+
"loss": 0.6398,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.15783841626538256,
|
426 |
+
"grad_norm": 2.6829258576996553,
|
427 |
+
"learning_rate": 9.898617700716988e-06,
|
428 |
+
"loss": 0.6721,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.16051364365971107,
|
433 |
+
"grad_norm": 2.365791638696942,
|
434 |
+
"learning_rate": 9.889048826294527e-06,
|
435 |
+
"loss": 0.6501,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.1631888710540396,
|
440 |
+
"grad_norm": 2.6788731378294535,
|
441 |
+
"learning_rate": 9.879053559929556e-06,
|
442 |
+
"loss": 0.6633,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.1658640984483681,
|
447 |
+
"grad_norm": 2.655949788782083,
|
448 |
+
"learning_rate": 9.868632773346044e-06,
|
449 |
+
"loss": 0.6512,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.16853932584269662,
|
454 |
+
"grad_norm": 2.525125627328949,
|
455 |
+
"learning_rate": 9.857787375379144e-06,
|
456 |
+
"loss": 0.6541,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.17121455323702514,
|
461 |
+
"grad_norm": 2.467491746753525,
|
462 |
+
"learning_rate": 9.84651831189593e-06,
|
463 |
+
"loss": 0.6579,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.17388978063135366,
|
468 |
+
"grad_norm": 2.516277938731903,
|
469 |
+
"learning_rate": 9.834826565712901e-06,
|
470 |
+
"loss": 0.6577,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.17656500802568217,
|
475 |
+
"grad_norm": 2.397791799699944,
|
476 |
+
"learning_rate": 9.822713156510278e-06,
|
477 |
+
"loss": 0.6496,
|
478 |
+
"step": 330
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.1792402354200107,
|
482 |
+
"grad_norm": 2.5057966279445854,
|
483 |
+
"learning_rate": 9.81017914074306e-06,
|
484 |
+
"loss": 0.6517,
|
485 |
+
"step": 335
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.1819154628143392,
|
489 |
+
"grad_norm": 2.4923621059461105,
|
490 |
+
"learning_rate": 9.797225611548896e-06,
|
491 |
+
"loss": 0.6688,
|
492 |
+
"step": 340
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.18459069020866772,
|
496 |
+
"grad_norm": 2.5049630606580613,
|
497 |
+
"learning_rate": 9.783853698652737e-06,
|
498 |
+
"loss": 0.6607,
|
499 |
+
"step": 345
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.18726591760299627,
|
503 |
+
"grad_norm": 2.358221426077587,
|
504 |
+
"learning_rate": 9.770064568268329e-06,
|
505 |
+
"loss": 0.652,
|
506 |
+
"step": 350
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.18994114499732478,
|
510 |
+
"grad_norm": 2.6263829483361367,
|
511 |
+
"learning_rate": 9.75585942299648e-06,
|
512 |
+
"loss": 0.6447,
|
513 |
+
"step": 355
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.1926163723916533,
|
517 |
+
"grad_norm": 2.4355704032991707,
|
518 |
+
"learning_rate": 9.741239501720197e-06,
|
519 |
+
"loss": 0.6447,
|
520 |
+
"step": 360
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.19529159978598182,
|
524 |
+
"grad_norm": 2.4808141511939796,
|
525 |
+
"learning_rate": 9.726206079496619e-06,
|
526 |
+
"loss": 0.6562,
|
527 |
+
"step": 365
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.19796682718031033,
|
531 |
+
"grad_norm": 2.7306912925058127,
|
532 |
+
"learning_rate": 9.71076046744583e-06,
|
533 |
+
"loss": 0.6355,
|
534 |
+
"step": 370
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.20064205457463885,
|
538 |
+
"grad_norm": 2.4644682995293317,
|
539 |
+
"learning_rate": 9.694904012636509e-06,
|
540 |
+
"loss": 0.6333,
|
541 |
+
"step": 375
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.20331728196896737,
|
545 |
+
"grad_norm": 2.4720584580787865,
|
546 |
+
"learning_rate": 9.678638097968435e-06,
|
547 |
+
"loss": 0.6195,
|
548 |
+
"step": 380
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.20599250936329588,
|
552 |
+
"grad_norm": 2.3633575413378503,
|
553 |
+
"learning_rate": 9.661964142051896e-06,
|
554 |
+
"loss": 0.6409,
|
555 |
+
"step": 385
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.2086677367576244,
|
559 |
+
"grad_norm": 2.3057026800286713,
|
560 |
+
"learning_rate": 9.644883599083959e-06,
|
561 |
+
"loss": 0.6246,
|
562 |
+
"step": 390
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.21134296415195292,
|
566 |
+
"grad_norm": 2.4532088121133246,
|
567 |
+
"learning_rate": 9.627397958721638e-06,
|
568 |
+
"loss": 0.642,
|
569 |
+
"step": 395
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.21401819154628143,
|
573 |
+
"grad_norm": 2.332423943377658,
|
574 |
+
"learning_rate": 9.609508745951988e-06,
|
575 |
+
"loss": 0.6226,
|
576 |
+
"step": 400
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.21669341894060995,
|
580 |
+
"grad_norm": 2.4978626441486833,
|
581 |
+
"learning_rate": 9.591217520959095e-06,
|
582 |
+
"loss": 0.6133,
|
583 |
+
"step": 405
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.21936864633493847,
|
587 |
+
"grad_norm": 2.2954192266671383,
|
588 |
+
"learning_rate": 9.572525878988014e-06,
|
589 |
+
"loss": 0.6148,
|
590 |
+
"step": 410
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.22204387372926698,
|
594 |
+
"grad_norm": 2.8095485182952564,
|
595 |
+
"learning_rate": 9.55343545020564e-06,
|
596 |
+
"loss": 0.6268,
|
597 |
+
"step": 415
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.2247191011235955,
|
601 |
+
"grad_norm": 2.3665476960634195,
|
602 |
+
"learning_rate": 9.533947899558521e-06,
|
603 |
+
"loss": 0.6244,
|
604 |
+
"step": 420
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.22739432851792402,
|
608 |
+
"grad_norm": 2.6753913761867523,
|
609 |
+
"learning_rate": 9.514064926627684e-06,
|
610 |
+
"loss": 0.6323,
|
611 |
+
"step": 425
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.23006955591225253,
|
615 |
+
"grad_norm": 2.5112745956277926,
|
616 |
+
"learning_rate": 9.49378826548037e-06,
|
617 |
+
"loss": 0.6195,
|
618 |
+
"step": 430
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.23274478330658105,
|
622 |
+
"grad_norm": 2.413468337663542,
|
623 |
+
"learning_rate": 9.473119684518834e-06,
|
624 |
+
"loss": 0.6074,
|
625 |
+
"step": 435
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.23542001070090957,
|
629 |
+
"grad_norm": 2.6553211159808567,
|
630 |
+
"learning_rate": 9.452060986326088e-06,
|
631 |
+
"loss": 0.6241,
|
632 |
+
"step": 440
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.23809523809523808,
|
636 |
+
"grad_norm": 2.659385798106578,
|
637 |
+
"learning_rate": 9.430614007508712e-06,
|
638 |
+
"loss": 0.6158,
|
639 |
+
"step": 445
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.24077046548956663,
|
643 |
+
"grad_norm": 2.2931792547740466,
|
644 |
+
"learning_rate": 9.408780618536664e-06,
|
645 |
+
"loss": 0.6018,
|
646 |
+
"step": 450
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.24344569288389514,
|
650 |
+
"grad_norm": 2.4514111664427043,
|
651 |
+
"learning_rate": 9.386562723580155e-06,
|
652 |
+
"loss": 0.6158,
|
653 |
+
"step": 455
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.24612092027822366,
|
657 |
+
"grad_norm": 2.420489212347836,
|
658 |
+
"learning_rate": 9.363962260343577e-06,
|
659 |
+
"loss": 0.5934,
|
660 |
+
"step": 460
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.24879614767255218,
|
664 |
+
"grad_norm": 2.5033351510386055,
|
665 |
+
"learning_rate": 9.340981199896515e-06,
|
666 |
+
"loss": 0.5992,
|
667 |
+
"step": 465
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.25147137506688066,
|
671 |
+
"grad_norm": 2.601071476421057,
|
672 |
+
"learning_rate": 9.317621546501827e-06,
|
673 |
+
"loss": 0.5921,
|
674 |
+
"step": 470
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.2541466024612092,
|
678 |
+
"grad_norm": 2.580094514637511,
|
679 |
+
"learning_rate": 9.293885337440869e-06,
|
680 |
+
"loss": 0.5966,
|
681 |
+
"step": 475
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.2568218298555377,
|
685 |
+
"grad_norm": 2.443217288458462,
|
686 |
+
"learning_rate": 9.26977464283579e-06,
|
687 |
+
"loss": 0.5945,
|
688 |
+
"step": 480
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.2594970572498662,
|
692 |
+
"grad_norm": 2.4281259437423177,
|
693 |
+
"learning_rate": 9.245291565469007e-06,
|
694 |
+
"loss": 0.5974,
|
695 |
+
"step": 485
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.26217228464419473,
|
699 |
+
"grad_norm": 2.6683703968935135,
|
700 |
+
"learning_rate": 9.220438240599813e-06,
|
701 |
+
"loss": 0.6115,
|
702 |
+
"step": 490
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.26484751203852325,
|
706 |
+
"grad_norm": 2.447016408813491,
|
707 |
+
"learning_rate": 9.19521683577814e-06,
|
708 |
+
"loss": 0.5878,
|
709 |
+
"step": 495
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.2675227394328518,
|
713 |
+
"grad_norm": 2.355863764903897,
|
714 |
+
"learning_rate": 9.169629550655532e-06,
|
715 |
+
"loss": 0.5758,
|
716 |
+
"step": 500
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.27019796682718034,
|
720 |
+
"grad_norm": 2.4204699917498806,
|
721 |
+
"learning_rate": 9.143678616793299e-06,
|
722 |
+
"loss": 0.5952,
|
723 |
+
"step": 505
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.27287319422150885,
|
727 |
+
"grad_norm": 2.4163211236401847,
|
728 |
+
"learning_rate": 9.117366297467899e-06,
|
729 |
+
"loss": 0.595,
|
730 |
+
"step": 510
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.27554842161583737,
|
734 |
+
"grad_norm": 2.502428814553625,
|
735 |
+
"learning_rate": 9.090694887473539e-06,
|
736 |
+
"loss": 0.5897,
|
737 |
+
"step": 515
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.2782236490101659,
|
741 |
+
"grad_norm": 2.568793263794682,
|
742 |
+
"learning_rate": 9.063666712922054e-06,
|
743 |
+
"loss": 0.5699,
|
744 |
+
"step": 520
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.2808988764044944,
|
748 |
+
"grad_norm": 2.382473697312008,
|
749 |
+
"learning_rate": 9.036284131040027e-06,
|
750 |
+
"loss": 0.589,
|
751 |
+
"step": 525
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.2835741037988229,
|
755 |
+
"grad_norm": 2.263455139276105,
|
756 |
+
"learning_rate": 9.008549529963202e-06,
|
757 |
+
"loss": 0.5864,
|
758 |
+
"step": 530
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.28624933119315144,
|
762 |
+
"grad_norm": 2.4693065814392527,
|
763 |
+
"learning_rate": 8.98046532852822e-06,
|
764 |
+
"loss": 0.5667,
|
765 |
+
"step": 535
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.28892455858747995,
|
769 |
+
"grad_norm": 2.4610990146399003,
|
770 |
+
"learning_rate": 8.952033976061651e-06,
|
771 |
+
"loss": 0.5978,
|
772 |
+
"step": 540
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.29159978598180847,
|
776 |
+
"grad_norm": 2.642358939686009,
|
777 |
+
"learning_rate": 8.923257952166391e-06,
|
778 |
+
"loss": 0.5821,
|
779 |
+
"step": 545
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.294275013376137,
|
783 |
+
"grad_norm": 2.5217057546849087,
|
784 |
+
"learning_rate": 8.894139766505391e-06,
|
785 |
+
"loss": 0.5793,
|
786 |
+
"step": 550
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.2969502407704655,
|
790 |
+
"grad_norm": 2.2661838032050468,
|
791 |
+
"learning_rate": 8.864681958582795e-06,
|
792 |
+
"loss": 0.567,
|
793 |
+
"step": 555
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.299625468164794,
|
797 |
+
"grad_norm": 2.442984986637873,
|
798 |
+
"learning_rate": 8.834887097522452e-06,
|
799 |
+
"loss": 0.558,
|
800 |
+
"step": 560
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.30230069555912253,
|
804 |
+
"grad_norm": 2.4181112148582184,
|
805 |
+
"learning_rate": 8.80475778184386e-06,
|
806 |
+
"loss": 0.5797,
|
807 |
+
"step": 565
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.30497592295345105,
|
811 |
+
"grad_norm": 2.4733057765728694,
|
812 |
+
"learning_rate": 8.774296639235527e-06,
|
813 |
+
"loss": 0.5729,
|
814 |
+
"step": 570
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.30765115034777957,
|
818 |
+
"grad_norm": 2.4046309203890264,
|
819 |
+
"learning_rate": 8.743506326325814e-06,
|
820 |
+
"loss": 0.5725,
|
821 |
+
"step": 575
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.3103263777421081,
|
825 |
+
"grad_norm": 2.3380230010121177,
|
826 |
+
"learning_rate": 8.712389528451236e-06,
|
827 |
+
"loss": 0.5461,
|
828 |
+
"step": 580
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.3130016051364366,
|
832 |
+
"grad_norm": 2.5038468197885835,
|
833 |
+
"learning_rate": 8.680948959422266e-06,
|
834 |
+
"loss": 0.5785,
|
835 |
+
"step": 585
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.3156768325307651,
|
839 |
+
"grad_norm": 3.1445703294824767,
|
840 |
+
"learning_rate": 8.649187361286641e-06,
|
841 |
+
"loss": 0.568,
|
842 |
+
"step": 590
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.31835205992509363,
|
846 |
+
"grad_norm": 2.406292242894352,
|
847 |
+
"learning_rate": 8.617107504090239e-06,
|
848 |
+
"loss": 0.557,
|
849 |
+
"step": 595
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.32102728731942215,
|
853 |
+
"grad_norm": 2.370295664665951,
|
854 |
+
"learning_rate": 8.584712185635477e-06,
|
855 |
+
"loss": 0.5601,
|
856 |
+
"step": 600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.32370251471375067,
|
860 |
+
"grad_norm": 2.7526828810884303,
|
861 |
+
"learning_rate": 8.552004231237308e-06,
|
862 |
+
"loss": 0.5592,
|
863 |
+
"step": 605
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.3263777421080792,
|
867 |
+
"grad_norm": 2.5217279078466603,
|
868 |
+
"learning_rate": 8.518986493476819e-06,
|
869 |
+
"loss": 0.5424,
|
870 |
+
"step": 610
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.3290529695024077,
|
874 |
+
"grad_norm": 2.4256807544144117,
|
875 |
+
"learning_rate": 8.485661851952443e-06,
|
876 |
+
"loss": 0.5558,
|
877 |
+
"step": 615
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.3317281968967362,
|
881 |
+
"grad_norm": 2.3341524059148084,
|
882 |
+
"learning_rate": 8.452033213028822e-06,
|
883 |
+
"loss": 0.5621,
|
884 |
+
"step": 620
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.33440342429106473,
|
888 |
+
"grad_norm": 2.3741670788238065,
|
889 |
+
"learning_rate": 8.418103509583323e-06,
|
890 |
+
"loss": 0.5625,
|
891 |
+
"step": 625
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.33707865168539325,
|
895 |
+
"grad_norm": 2.3488707698974465,
|
896 |
+
"learning_rate": 8.383875700750272e-06,
|
897 |
+
"loss": 0.5564,
|
898 |
+
"step": 630
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.33975387907972177,
|
902 |
+
"grad_norm": 2.2944826885447456,
|
903 |
+
"learning_rate": 8.349352771662848e-06,
|
904 |
+
"loss": 0.5448,
|
905 |
+
"step": 635
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.3424291064740503,
|
909 |
+
"grad_norm": 2.340765525681543,
|
910 |
+
"learning_rate": 8.314537733192762e-06,
|
911 |
+
"loss": 0.5464,
|
912 |
+
"step": 640
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.3451043338683788,
|
916 |
+
"grad_norm": 2.4912590904296383,
|
917 |
+
"learning_rate": 8.279433621687658e-06,
|
918 |
+
"loss": 0.5508,
|
919 |
+
"step": 645
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.3477795612627073,
|
923 |
+
"grad_norm": 2.4869220531713974,
|
924 |
+
"learning_rate": 8.2440434987063e-06,
|
925 |
+
"loss": 0.539,
|
926 |
+
"step": 650
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.35045478865703583,
|
930 |
+
"grad_norm": 2.3322530356260986,
|
931 |
+
"learning_rate": 8.208370450751568e-06,
|
932 |
+
"loss": 0.5245,
|
933 |
+
"step": 655
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.35313001605136435,
|
937 |
+
"grad_norm": 2.4501780549096113,
|
938 |
+
"learning_rate": 8.172417589001275e-06,
|
939 |
+
"loss": 0.5458,
|
940 |
+
"step": 660
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.35580524344569286,
|
944 |
+
"grad_norm": 2.429282446394144,
|
945 |
+
"learning_rate": 8.136188049036817e-06,
|
946 |
+
"loss": 0.5457,
|
947 |
+
"step": 665
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.3584804708400214,
|
951 |
+
"grad_norm": 2.5498468100287046,
|
952 |
+
"learning_rate": 8.099684990569717e-06,
|
953 |
+
"loss": 0.5253,
|
954 |
+
"step": 670
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.3611556982343499,
|
958 |
+
"grad_norm": 2.3702949455068683,
|
959 |
+
"learning_rate": 8.06291159716606e-06,
|
960 |
+
"loss": 0.5426,
|
961 |
+
"step": 675
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.3638309256286784,
|
965 |
+
"grad_norm": 2.546455034570349,
|
966 |
+
"learning_rate": 8.025871075968828e-06,
|
967 |
+
"loss": 0.5285,
|
968 |
+
"step": 680
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.36650615302300693,
|
972 |
+
"grad_norm": 2.286543746187905,
|
973 |
+
"learning_rate": 7.988566657418202e-06,
|
974 |
+
"loss": 0.5249,
|
975 |
+
"step": 685
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.36918138041733545,
|
979 |
+
"grad_norm": 2.3795001406889416,
|
980 |
+
"learning_rate": 7.951001594969827e-06,
|
981 |
+
"loss": 0.5382,
|
982 |
+
"step": 690
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.37185660781166396,
|
986 |
+
"grad_norm": 2.5367851021499273,
|
987 |
+
"learning_rate": 7.91317916481106e-06,
|
988 |
+
"loss": 0.5313,
|
989 |
+
"step": 695
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.37453183520599254,
|
993 |
+
"grad_norm": 2.4790821311175866,
|
994 |
+
"learning_rate": 7.875102665575241e-06,
|
995 |
+
"loss": 0.5364,
|
996 |
+
"step": 700
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.37720706260032105,
|
1000 |
+
"grad_norm": 2.436587808572014,
|
1001 |
+
"learning_rate": 7.83677541805401e-06,
|
1002 |
+
"loss": 0.5285,
|
1003 |
+
"step": 705
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.37988228999464957,
|
1007 |
+
"grad_norm": 2.37913475078827,
|
1008 |
+
"learning_rate": 7.798200764907691e-06,
|
1009 |
+
"loss": 0.5199,
|
1010 |
+
"step": 710
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.3825575173889781,
|
1014 |
+
"grad_norm": 2.5005240559845134,
|
1015 |
+
"learning_rate": 7.759382070373755e-06,
|
1016 |
+
"loss": 0.5358,
|
1017 |
+
"step": 715
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.3852327447833066,
|
1021 |
+
"grad_norm": 2.4480000456926625,
|
1022 |
+
"learning_rate": 7.720322719973433e-06,
|
1023 |
+
"loss": 0.5344,
|
1024 |
+
"step": 720
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.3879079721776351,
|
1028 |
+
"grad_norm": 2.2698845818992455,
|
1029 |
+
"learning_rate": 7.68102612021643e-06,
|
1030 |
+
"loss": 0.5279,
|
1031 |
+
"step": 725
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.39058319957196364,
|
1035 |
+
"grad_norm": 2.4186655398319252,
|
1036 |
+
"learning_rate": 7.641495698303844e-06,
|
1037 |
+
"loss": 0.5169,
|
1038 |
+
"step": 730
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.39325842696629215,
|
1042 |
+
"grad_norm": 2.5271248633702674,
|
1043 |
+
"learning_rate": 7.60173490182926e-06,
|
1044 |
+
"loss": 0.5232,
|
1045 |
+
"step": 735
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.39593365436062067,
|
1049 |
+
"grad_norm": 2.2703784304673005,
|
1050 |
+
"learning_rate": 7.5617471984780885e-06,
|
1051 |
+
"loss": 0.528,
|
1052 |
+
"step": 740
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.3986088817549492,
|
1056 |
+
"grad_norm": 2.532874522915725,
|
1057 |
+
"learning_rate": 7.521536075725106e-06,
|
1058 |
+
"loss": 0.5249,
|
1059 |
+
"step": 745
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.4012841091492777,
|
1063 |
+
"grad_norm": 2.4292067983633077,
|
1064 |
+
"learning_rate": 7.481105040530334e-06,
|
1065 |
+
"loss": 0.5369,
|
1066 |
+
"step": 750
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.4039593365436062,
|
1070 |
+
"grad_norm": 2.3933411598642698,
|
1071 |
+
"learning_rate": 7.440457619033155e-06,
|
1072 |
+
"loss": 0.5205,
|
1073 |
+
"step": 755
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.40663456393793473,
|
1077 |
+
"grad_norm": 2.6309076094473935,
|
1078 |
+
"learning_rate": 7.3995973562448065e-06,
|
1079 |
+
"loss": 0.4936,
|
1080 |
+
"step": 760
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.40930979133226325,
|
1084 |
+
"grad_norm": 2.4382597285863916,
|
1085 |
+
"learning_rate": 7.358527815739192e-06,
|
1086 |
+
"loss": 0.5397,
|
1087 |
+
"step": 765
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.41198501872659177,
|
1091 |
+
"grad_norm": 2.493128592305897,
|
1092 |
+
"learning_rate": 7.317252579342096e-06,
|
1093 |
+
"loss": 0.5062,
|
1094 |
+
"step": 770
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.4146602461209203,
|
1098 |
+
"grad_norm": 2.39571739461516,
|
1099 |
+
"learning_rate": 7.275775246818802e-06,
|
1100 |
+
"loss": 0.5077,
|
1101 |
+
"step": 775
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.4173354735152488,
|
1105 |
+
"grad_norm": 2.3257482836528798,
|
1106 |
+
"learning_rate": 7.23409943556014e-06,
|
1107 |
+
"loss": 0.5051,
|
1108 |
+
"step": 780
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.4200107009095773,
|
1112 |
+
"grad_norm": 2.3506716083497623,
|
1113 |
+
"learning_rate": 7.192228780266997e-06,
|
1114 |
+
"loss": 0.5051,
|
1115 |
+
"step": 785
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.42268592830390583,
|
1119 |
+
"grad_norm": 2.3896789190082948,
|
1120 |
+
"learning_rate": 7.150166932633328e-06,
|
1121 |
+
"loss": 0.4925,
|
1122 |
+
"step": 790
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.42536115569823435,
|
1126 |
+
"grad_norm": 2.304737494822103,
|
1127 |
+
"learning_rate": 7.1079175610276775e-06,
|
1128 |
+
"loss": 0.516,
|
1129 |
+
"step": 795
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.42803638309256287,
|
1133 |
+
"grad_norm": 2.376368349197412,
|
1134 |
+
"learning_rate": 7.065484350173242e-06,
|
1135 |
+
"loss": 0.508,
|
1136 |
+
"step": 800
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.4307116104868914,
|
1140 |
+
"grad_norm": 2.390599123508061,
|
1141 |
+
"learning_rate": 7.022871000826519e-06,
|
1142 |
+
"loss": 0.4957,
|
1143 |
+
"step": 805
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.4333868378812199,
|
1147 |
+
"grad_norm": 2.4503319503961496,
|
1148 |
+
"learning_rate": 6.980081229454545e-06,
|
1149 |
+
"loss": 0.4932,
|
1150 |
+
"step": 810
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.4360620652755484,
|
1154 |
+
"grad_norm": 2.388736493290777,
|
1155 |
+
"learning_rate": 6.937118767910771e-06,
|
1156 |
+
"loss": 0.5031,
|
1157 |
+
"step": 815
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.43873729266987693,
|
1161 |
+
"grad_norm": 2.34192919036682,
|
1162 |
+
"learning_rate": 6.893987363109595e-06,
|
1163 |
+
"loss": 0.4964,
|
1164 |
+
"step": 820
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.44141252006420545,
|
1168 |
+
"grad_norm": 2.525102607168378,
|
1169 |
+
"learning_rate": 6.850690776699574e-06,
|
1170 |
+
"loss": 0.4976,
|
1171 |
+
"step": 825
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.44408774745853397,
|
1175 |
+
"grad_norm": 2.582188474917445,
|
1176 |
+
"learning_rate": 6.807232784735363e-06,
|
1177 |
+
"loss": 0.4798,
|
1178 |
+
"step": 830
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.4467629748528625,
|
1182 |
+
"grad_norm": 2.3217690049430937,
|
1183 |
+
"learning_rate": 6.763617177348394e-06,
|
1184 |
+
"loss": 0.4882,
|
1185 |
+
"step": 835
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.449438202247191,
|
1189 |
+
"grad_norm": 2.3691948839930643,
|
1190 |
+
"learning_rate": 6.719847758416316e-06,
|
1191 |
+
"loss": 0.4921,
|
1192 |
+
"step": 840
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.4521134296415195,
|
1196 |
+
"grad_norm": 2.384102404640162,
|
1197 |
+
"learning_rate": 6.675928345231248e-06,
|
1198 |
+
"loss": 0.4872,
|
1199 |
+
"step": 845
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.45478865703584803,
|
1203 |
+
"grad_norm": 2.396616687246066,
|
1204 |
+
"learning_rate": 6.631862768166861e-06,
|
1205 |
+
"loss": 0.4982,
|
1206 |
+
"step": 850
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.45746388443017655,
|
1210 |
+
"grad_norm": 2.493144015288168,
|
1211 |
+
"learning_rate": 6.587654870344318e-06,
|
1212 |
+
"loss": 0.4927,
|
1213 |
+
"step": 855
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.46013911182450506,
|
1217 |
+
"grad_norm": 2.4553160494373296,
|
1218 |
+
"learning_rate": 6.543308507297094e-06,
|
1219 |
+
"loss": 0.4924,
|
1220 |
+
"step": 860
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.4628143392188336,
|
1224 |
+
"grad_norm": 2.3226795438252354,
|
1225 |
+
"learning_rate": 6.498827546634733e-06,
|
1226 |
+
"loss": 0.4709,
|
1227 |
+
"step": 865
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.4654895666131621,
|
1231 |
+
"grad_norm": 2.281005549264743,
|
1232 |
+
"learning_rate": 6.454215867705526e-06,
|
1233 |
+
"loss": 0.4983,
|
1234 |
+
"step": 870
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.4681647940074906,
|
1238 |
+
"grad_norm": 2.4555385388069473,
|
1239 |
+
"learning_rate": 6.409477361258188e-06,
|
1240 |
+
"loss": 0.4753,
|
1241 |
+
"step": 875
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.47084002140181913,
|
1245 |
+
"grad_norm": 2.4255588194409583,
|
1246 |
+
"learning_rate": 6.364615929102531e-06,
|
1247 |
+
"loss": 0.4763,
|
1248 |
+
"step": 880
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.47351524879614765,
|
1252 |
+
"grad_norm": 2.406190699304918,
|
1253 |
+
"learning_rate": 6.319635483769164e-06,
|
1254 |
+
"loss": 0.4843,
|
1255 |
+
"step": 885
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.47619047619047616,
|
1259 |
+
"grad_norm": 2.386893315693651,
|
1260 |
+
"learning_rate": 6.274539948168279e-06,
|
1261 |
+
"loss": 0.4716,
|
1262 |
+
"step": 890
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.47886570358480474,
|
1266 |
+
"grad_norm": 2.2728664616613887,
|
1267 |
+
"learning_rate": 6.229333255247511e-06,
|
1268 |
+
"loss": 0.467,
|
1269 |
+
"step": 895
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.48154093097913325,
|
1273 |
+
"grad_norm": 2.4446705313074064,
|
1274 |
+
"learning_rate": 6.184019347648939e-06,
|
1275 |
+
"loss": 0.4683,
|
1276 |
+
"step": 900
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.48421615837346177,
|
1280 |
+
"grad_norm": 2.297929824104002,
|
1281 |
+
"learning_rate": 6.138602177365218e-06,
|
1282 |
+
"loss": 0.4814,
|
1283 |
+
"step": 905
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.4868913857677903,
|
1287 |
+
"grad_norm": 2.2398605800909226,
|
1288 |
+
"learning_rate": 6.093085705394934e-06,
|
1289 |
+
"loss": 0.4761,
|
1290 |
+
"step": 910
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.4895666131621188,
|
1294 |
+
"grad_norm": 2.4670515565573083,
|
1295 |
+
"learning_rate": 6.04747390139713e-06,
|
1296 |
+
"loss": 0.4705,
|
1297 |
+
"step": 915
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.4922418405564473,
|
1301 |
+
"grad_norm": 2.436162832733763,
|
1302 |
+
"learning_rate": 6.001770743345108e-06,
|
1303 |
+
"loss": 0.4689,
|
1304 |
+
"step": 920
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.49491706795077584,
|
1308 |
+
"grad_norm": 2.7004239367895053,
|
1309 |
+
"learning_rate": 5.9559802171794955e-06,
|
1310 |
+
"loss": 0.464,
|
1311 |
+
"step": 925
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.49759229534510435,
|
1315 |
+
"grad_norm": 2.352197390213324,
|
1316 |
+
"learning_rate": 5.9101063164606165e-06,
|
1317 |
+
"loss": 0.4588,
|
1318 |
+
"step": 930
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.5002675227394329,
|
1322 |
+
"grad_norm": 2.386178979724787,
|
1323 |
+
"learning_rate": 5.864153042020191e-06,
|
1324 |
+
"loss": 0.4593,
|
1325 |
+
"step": 935
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.5029427501337613,
|
1329 |
+
"grad_norm": 2.271248403727243,
|
1330 |
+
"learning_rate": 5.818124401612416e-06,
|
1331 |
+
"loss": 0.4554,
|
1332 |
+
"step": 940
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.5056179775280899,
|
1336 |
+
"grad_norm": 2.34018371964795,
|
1337 |
+
"learning_rate": 5.7720244095644305e-06,
|
1338 |
+
"loss": 0.4703,
|
1339 |
+
"step": 945
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.5082932049224184,
|
1343 |
+
"grad_norm": 2.270516604487789,
|
1344 |
+
"learning_rate": 5.725857086426216e-06,
|
1345 |
+
"loss": 0.454,
|
1346 |
+
"step": 950
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.5109684323167469,
|
1350 |
+
"grad_norm": 2.3872382871657325,
|
1351 |
+
"learning_rate": 5.679626458619947e-06,
|
1352 |
+
"loss": 0.4457,
|
1353 |
+
"step": 955
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.5136436597110754,
|
1357 |
+
"grad_norm": 2.520375771831823,
|
1358 |
+
"learning_rate": 5.633336558088823e-06,
|
1359 |
+
"loss": 0.4598,
|
1360 |
+
"step": 960
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.516318887105404,
|
1364 |
+
"grad_norm": 2.2353434501523575,
|
1365 |
+
"learning_rate": 5.586991421945445e-06,
|
1366 |
+
"loss": 0.4711,
|
1367 |
+
"step": 965
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.5189941144997324,
|
1371 |
+
"grad_norm": 2.2812982114304563,
|
1372 |
+
"learning_rate": 5.540595092119709e-06,
|
1373 |
+
"loss": 0.4439,
|
1374 |
+
"step": 970
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.521669341894061,
|
1378 |
+
"grad_norm": 2.392550181275633,
|
1379 |
+
"learning_rate": 5.494151615006307e-06,
|
1380 |
+
"loss": 0.4604,
|
1381 |
+
"step": 975
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.5243445692883895,
|
1385 |
+
"grad_norm": 2.233410991408843,
|
1386 |
+
"learning_rate": 5.44766504111181e-06,
|
1387 |
+
"loss": 0.4674,
|
1388 |
+
"step": 980
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.527019796682718,
|
1392 |
+
"grad_norm": 2.4198668089222153,
|
1393 |
+
"learning_rate": 5.401139424701427e-06,
|
1394 |
+
"loss": 0.4553,
|
1395 |
+
"step": 985
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.5296950240770465,
|
1399 |
+
"grad_norm": 2.377473414386296,
|
1400 |
+
"learning_rate": 5.354578823445404e-06,
|
1401 |
+
"loss": 0.4477,
|
1402 |
+
"step": 990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.5323702514713751,
|
1406 |
+
"grad_norm": 2.3622932708922213,
|
1407 |
+
"learning_rate": 5.307987298065145e-06,
|
1408 |
+
"loss": 0.4339,
|
1409 |
+
"step": 995
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.5350454788657036,
|
1413 |
+
"grad_norm": 2.332131787995006,
|
1414 |
+
"learning_rate": 5.26136891197906e-06,
|
1415 |
+
"loss": 0.4404,
|
1416 |
+
"step": 1000
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.5377207062600321,
|
1420 |
+
"grad_norm": 2.4251937487497783,
|
1421 |
+
"learning_rate": 5.214727730948181e-06,
|
1422 |
+
"loss": 0.4566,
|
1423 |
+
"step": 1005
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.5403959336543607,
|
1427 |
+
"grad_norm": 2.3957788984396937,
|
1428 |
+
"learning_rate": 5.1680678227215705e-06,
|
1429 |
+
"loss": 0.4405,
|
1430 |
+
"step": 1010
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.5430711610486891,
|
1434 |
+
"grad_norm": 2.3847750525666873,
|
1435 |
+
"learning_rate": 5.121393256681561e-06,
|
1436 |
+
"loss": 0.4446,
|
1437 |
+
"step": 1015
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.5457463884430177,
|
1441 |
+
"grad_norm": 2.2698162072049533,
|
1442 |
+
"learning_rate": 5.07470810348884e-06,
|
1443 |
+
"loss": 0.4354,
|
1444 |
+
"step": 1020
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.5484216158373462,
|
1448 |
+
"grad_norm": 2.436311734783881,
|
1449 |
+
"learning_rate": 5.02801643472745e-06,
|
1450 |
+
"loss": 0.4414,
|
1451 |
+
"step": 1025
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.5510968432316747,
|
1455 |
+
"grad_norm": 2.2916945959636643,
|
1456 |
+
"learning_rate": 4.98132232254967e-06,
|
1457 |
+
"loss": 0.4391,
|
1458 |
+
"step": 1030
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.5537720706260032,
|
1462 |
+
"grad_norm": 2.4013743839662602,
|
1463 |
+
"learning_rate": 4.934629839320885e-06,
|
1464 |
+
"loss": 0.4371,
|
1465 |
+
"step": 1035
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.5564472980203318,
|
1469 |
+
"grad_norm": 2.3988828791639314,
|
1470 |
+
"learning_rate": 4.88794305726441e-06,
|
1471 |
+
"loss": 0.4375,
|
1472 |
+
"step": 1040
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.5591225254146602,
|
1476 |
+
"grad_norm": 2.330590397904805,
|
1477 |
+
"learning_rate": 4.841266048106343e-06,
|
1478 |
+
"loss": 0.434,
|
1479 |
+
"step": 1045
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.5617977528089888,
|
1483 |
+
"grad_norm": 2.3366142392042124,
|
1484 |
+
"learning_rate": 4.794602882720448e-06,
|
1485 |
+
"loss": 0.4281,
|
1486 |
+
"step": 1050
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.5644729802033173,
|
1490 |
+
"grad_norm": 2.3883999122895503,
|
1491 |
+
"learning_rate": 4.747957630773124e-06,
|
1492 |
+
"loss": 0.4317,
|
1493 |
+
"step": 1055
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.5671482075976458,
|
1497 |
+
"grad_norm": 2.3510366869048323,
|
1498 |
+
"learning_rate": 4.701334360368473e-06,
|
1499 |
+
"loss": 0.4272,
|
1500 |
+
"step": 1060
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.5698234349919743,
|
1504 |
+
"grad_norm": 2.3683109214556315,
|
1505 |
+
"learning_rate": 4.654737137693508e-06,
|
1506 |
+
"loss": 0.4381,
|
1507 |
+
"step": 1065
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.5724986623863029,
|
1511 |
+
"grad_norm": 2.2171740005409304,
|
1512 |
+
"learning_rate": 4.6081700266635195e-06,
|
1513 |
+
"loss": 0.4227,
|
1514 |
+
"step": 1070
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.5751738897806313,
|
1518 |
+
"grad_norm": 2.3232281982068783,
|
1519 |
+
"learning_rate": 4.561637088567654e-06,
|
1520 |
+
"loss": 0.4217,
|
1521 |
+
"step": 1075
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.5778491171749599,
|
1525 |
+
"grad_norm": 2.4194928745884514,
|
1526 |
+
"learning_rate": 4.51514238171471e-06,
|
1527 |
+
"loss": 0.4255,
|
1528 |
+
"step": 1080
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.5805243445692884,
|
1532 |
+
"grad_norm": 2.202777210726338,
|
1533 |
+
"learning_rate": 4.468689961079195e-06,
|
1534 |
+
"loss": 0.4327,
|
1535 |
+
"step": 1085
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.5831995719636169,
|
1539 |
+
"grad_norm": 2.2294786297275926,
|
1540 |
+
"learning_rate": 4.4222838779476866e-06,
|
1541 |
+
"loss": 0.4352,
|
1542 |
+
"step": 1090
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.5858747993579454,
|
1546 |
+
"grad_norm": 2.3255340067920933,
|
1547 |
+
"learning_rate": 4.375928179565494e-06,
|
1548 |
+
"loss": 0.418,
|
1549 |
+
"step": 1095
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.588550026752274,
|
1553 |
+
"grad_norm": 2.3996404743348205,
|
1554 |
+
"learning_rate": 4.329626908783685e-06,
|
1555 |
+
"loss": 0.4189,
|
1556 |
+
"step": 1100
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.5912252541466024,
|
1560 |
+
"grad_norm": 2.303291302129872,
|
1561 |
+
"learning_rate": 4.2833841037065e-06,
|
1562 |
+
"loss": 0.4077,
|
1563 |
+
"step": 1105
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.593900481540931,
|
1567 |
+
"grad_norm": 2.301388594139086,
|
1568 |
+
"learning_rate": 4.237203797339169e-06,
|
1569 |
+
"loss": 0.4104,
|
1570 |
+
"step": 1110
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.5965757089352595,
|
1574 |
+
"grad_norm": 2.160810901865994,
|
1575 |
+
"learning_rate": 4.191090017236177e-06,
|
1576 |
+
"loss": 0.4235,
|
1577 |
+
"step": 1115
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.599250936329588,
|
1581 |
+
"grad_norm": 2.2158867728636413,
|
1582 |
+
"learning_rate": 4.145046785150013e-06,
|
1583 |
+
"loss": 0.4046,
|
1584 |
+
"step": 1120
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.6019261637239165,
|
1588 |
+
"grad_norm": 2.2735600817905923,
|
1589 |
+
"learning_rate": 4.09907811668041e-06,
|
1590 |
+
"loss": 0.408,
|
1591 |
+
"step": 1125
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.6046013911182451,
|
1595 |
+
"grad_norm": 2.3803804980296444,
|
1596 |
+
"learning_rate": 4.0531880209241356e-06,
|
1597 |
+
"loss": 0.4136,
|
1598 |
+
"step": 1130
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.6072766185125735,
|
1602 |
+
"grad_norm": 2.266085861670805,
|
1603 |
+
"learning_rate": 4.0073805001253405e-06,
|
1604 |
+
"loss": 0.4051,
|
1605 |
+
"step": 1135
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.6099518459069021,
|
1609 |
+
"grad_norm": 2.1954654483630796,
|
1610 |
+
"learning_rate": 3.961659549326512e-06,
|
1611 |
+
"loss": 0.4262,
|
1612 |
+
"step": 1140
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.6126270733012306,
|
1616 |
+
"grad_norm": 2.228408225766588,
|
1617 |
+
"learning_rate": 3.916029156020044e-06,
|
1618 |
+
"loss": 0.4029,
|
1619 |
+
"step": 1145
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.6153023006955591,
|
1623 |
+
"grad_norm": 2.3199361261442144,
|
1624 |
+
"learning_rate": 3.870493299800484e-06,
|
1625 |
+
"loss": 0.4168,
|
1626 |
+
"step": 1150
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.6179775280898876,
|
1630 |
+
"grad_norm": 2.2619585349946796,
|
1631 |
+
"learning_rate": 3.82505595201745e-06,
|
1632 |
+
"loss": 0.408,
|
1633 |
+
"step": 1155
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.6206527554842162,
|
1637 |
+
"grad_norm": 2.4880856896093544,
|
1638 |
+
"learning_rate": 3.7797210754292766e-06,
|
1639 |
+
"loss": 0.4128,
|
1640 |
+
"step": 1160
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.6233279828785446,
|
1644 |
+
"grad_norm": 2.212702659150222,
|
1645 |
+
"learning_rate": 3.7344926238574074e-06,
|
1646 |
+
"loss": 0.4121,
|
1647 |
+
"step": 1165
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.6260032102728732,
|
1651 |
+
"grad_norm": 2.262392758862919,
|
1652 |
+
"learning_rate": 3.6893745418415692e-06,
|
1653 |
+
"loss": 0.3935,
|
1654 |
+
"step": 1170
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.6286784376672017,
|
1658 |
+
"grad_norm": 2.2203307618409185,
|
1659 |
+
"learning_rate": 3.6443707642957526e-06,
|
1660 |
+
"loss": 0.3996,
|
1661 |
+
"step": 1175
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.6313536650615302,
|
1665 |
+
"grad_norm": 2.183173739658585,
|
1666 |
+
"learning_rate": 3.5994852161650386e-06,
|
1667 |
+
"loss": 0.389,
|
1668 |
+
"step": 1180
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.6340288924558587,
|
1672 |
+
"grad_norm": 2.3980317993943814,
|
1673 |
+
"learning_rate": 3.5547218120832807e-06,
|
1674 |
+
"loss": 0.3936,
|
1675 |
+
"step": 1185
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.6367041198501873,
|
1679 |
+
"grad_norm": 2.3294327156692654,
|
1680 |
+
"learning_rate": 3.5100844560317028e-06,
|
1681 |
+
"loss": 0.4095,
|
1682 |
+
"step": 1190
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.6393793472445158,
|
1686 |
+
"grad_norm": 2.3318739063082936,
|
1687 |
+
"learning_rate": 3.465577040998417e-06,
|
1688 |
+
"loss": 0.4036,
|
1689 |
+
"step": 1195
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.6420545746388443,
|
1693 |
+
"grad_norm": 2.13554424918326,
|
1694 |
+
"learning_rate": 3.4212034486388972e-06,
|
1695 |
+
"loss": 0.385,
|
1696 |
+
"step": 1200
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.6447298020331729,
|
1700 |
+
"grad_norm": 2.3059775180831887,
|
1701 |
+
"learning_rate": 3.376967548937457e-06,
|
1702 |
+
"loss": 0.3959,
|
1703 |
+
"step": 1205
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.6474050294275013,
|
1707 |
+
"grad_norm": 2.2003509075548124,
|
1708 |
+
"learning_rate": 3.332873199869719e-06,
|
1709 |
+
"loss": 0.3918,
|
1710 |
+
"step": 1210
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.6500802568218299,
|
1714 |
+
"grad_norm": 2.2638155453662576,
|
1715 |
+
"learning_rate": 3.2889242470661553e-06,
|
1716 |
+
"loss": 0.3964,
|
1717 |
+
"step": 1215
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.6527554842161584,
|
1721 |
+
"grad_norm": 2.2328176385540126,
|
1722 |
+
"learning_rate": 3.245124523476699e-06,
|
1723 |
+
"loss": 0.3872,
|
1724 |
+
"step": 1220
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.6554307116104869,
|
1728 |
+
"grad_norm": 2.2612712629675364,
|
1729 |
+
"learning_rate": 3.2014778490364484e-06,
|
1730 |
+
"loss": 0.4019,
|
1731 |
+
"step": 1225
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.6581059390048154,
|
1735 |
+
"grad_norm": 2.183300299715255,
|
1736 |
+
"learning_rate": 3.157988030332526e-06,
|
1737 |
+
"loss": 0.3986,
|
1738 |
+
"step": 1230
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.660781166399144,
|
1742 |
+
"grad_norm": 2.2008867618272245,
|
1743 |
+
"learning_rate": 3.1146588602720884e-06,
|
1744 |
+
"loss": 0.3793,
|
1745 |
+
"step": 1235
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.6634563937934724,
|
1749 |
+
"grad_norm": 2.3131230110616396,
|
1750 |
+
"learning_rate": 3.0714941177515307e-06,
|
1751 |
+
"loss": 0.389,
|
1752 |
+
"step": 1240
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.666131621187801,
|
1756 |
+
"grad_norm": 2.3622281138792105,
|
1757 |
+
"learning_rate": 3.0284975673269175e-06,
|
1758 |
+
"loss": 0.3906,
|
1759 |
+
"step": 1245
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.6688068485821295,
|
1763 |
+
"grad_norm": 2.2764800784114696,
|
1764 |
+
"learning_rate": 2.9856729588856615e-06,
|
1765 |
+
"loss": 0.3917,
|
1766 |
+
"step": 1250
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.671482075976458,
|
1770 |
+
"grad_norm": 2.1993222173811064,
|
1771 |
+
"learning_rate": 2.9430240273194844e-06,
|
1772 |
+
"loss": 0.3829,
|
1773 |
+
"step": 1255
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.6741573033707865,
|
1777 |
+
"grad_norm": 2.2083725272220414,
|
1778 |
+
"learning_rate": 2.9005544921986774e-06,
|
1779 |
+
"loss": 0.3778,
|
1780 |
+
"step": 1260
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.6768325307651151,
|
1784 |
+
"grad_norm": 2.214779231475582,
|
1785 |
+
"learning_rate": 2.858268057447712e-06,
|
1786 |
+
"loss": 0.3901,
|
1787 |
+
"step": 1265
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.6795077581594435,
|
1791 |
+
"grad_norm": 2.2254433327654097,
|
1792 |
+
"learning_rate": 2.8161684110221987e-06,
|
1793 |
+
"loss": 0.387,
|
1794 |
+
"step": 1270
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.6821829855537721,
|
1798 |
+
"grad_norm": 2.31973974930244,
|
1799 |
+
"learning_rate": 2.7742592245872523e-06,
|
1800 |
+
"loss": 0.3852,
|
1801 |
+
"step": 1275
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.6848582129481006,
|
1805 |
+
"grad_norm": 2.212647002572668,
|
1806 |
+
"learning_rate": 2.7325441531972685e-06,
|
1807 |
+
"loss": 0.3753,
|
1808 |
+
"step": 1280
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.6875334403424291,
|
1812 |
+
"grad_norm": 2.1846100608702606,
|
1813 |
+
"learning_rate": 2.691026834977161e-06,
|
1814 |
+
"loss": 0.3995,
|
1815 |
+
"step": 1285
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.6902086677367576,
|
1819 |
+
"grad_norm": 2.229995641206547,
|
1820 |
+
"learning_rate": 2.649710890805055e-06,
|
1821 |
+
"loss": 0.3828,
|
1822 |
+
"step": 1290
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.6928838951310862,
|
1826 |
+
"grad_norm": 2.1422721805468354,
|
1827 |
+
"learning_rate": 2.6085999239965094e-06,
|
1828 |
+
"loss": 0.3692,
|
1829 |
+
"step": 1295
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.6955591225254146,
|
1833 |
+
"grad_norm": 2.318213911947721,
|
1834 |
+
"learning_rate": 2.567697519990249e-06,
|
1835 |
+
"loss": 0.3796,
|
1836 |
+
"step": 1300
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.6982343499197432,
|
1840 |
+
"grad_norm": 2.2119506235831827,
|
1841 |
+
"learning_rate": 2.52700724603547e-06,
|
1842 |
+
"loss": 0.3775,
|
1843 |
+
"step": 1305
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.7009095773140717,
|
1847 |
+
"grad_norm": 2.224473043074625,
|
1848 |
+
"learning_rate": 2.4865326508807274e-06,
|
1849 |
+
"loss": 0.37,
|
1850 |
+
"step": 1310
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.7035848047084002,
|
1854 |
+
"grad_norm": 2.079689857122922,
|
1855 |
+
"learning_rate": 2.446277264464431e-06,
|
1856 |
+
"loss": 0.3771,
|
1857 |
+
"step": 1315
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.7062600321027287,
|
1861 |
+
"grad_norm": 2.277862216469956,
|
1862 |
+
"learning_rate": 2.406244597606994e-06,
|
1863 |
+
"loss": 0.3673,
|
1864 |
+
"step": 1320
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.7089352594970573,
|
1868 |
+
"grad_norm": 2.1871262495624952,
|
1869 |
+
"learning_rate": 2.3664381417046362e-06,
|
1870 |
+
"loss": 0.372,
|
1871 |
+
"step": 1325
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.7116104868913857,
|
1875 |
+
"grad_norm": 2.5202258537627293,
|
1876 |
+
"learning_rate": 2.3268613684248846e-06,
|
1877 |
+
"loss": 0.381,
|
1878 |
+
"step": 1330
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.7142857142857143,
|
1882 |
+
"grad_norm": 2.2472346697981296,
|
1883 |
+
"learning_rate": 2.287517729403802e-06,
|
1884 |
+
"loss": 0.368,
|
1885 |
+
"step": 1335
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.7169609416800428,
|
1889 |
+
"grad_norm": 2.221291336010394,
|
1890 |
+
"learning_rate": 2.2484106559449527e-06,
|
1891 |
+
"loss": 0.363,
|
1892 |
+
"step": 1340
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.7196361690743713,
|
1896 |
+
"grad_norm": 2.183019746627642,
|
1897 |
+
"learning_rate": 2.2095435587201487e-06,
|
1898 |
+
"loss": 0.3705,
|
1899 |
+
"step": 1345
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.7223113964686998,
|
1903 |
+
"grad_norm": 2.1399656648022933,
|
1904 |
+
"learning_rate": 2.1709198274719908e-06,
|
1905 |
+
"loss": 0.3654,
|
1906 |
+
"step": 1350
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.7249866238630284,
|
1910 |
+
"grad_norm": 2.0861165370537105,
|
1911 |
+
"learning_rate": 2.1325428307182357e-06,
|
1912 |
+
"loss": 0.3669,
|
1913 |
+
"step": 1355
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.7276618512573568,
|
1917 |
+
"grad_norm": 2.164652924106045,
|
1918 |
+
"learning_rate": 2.0944159154580225e-06,
|
1919 |
+
"loss": 0.3696,
|
1920 |
+
"step": 1360
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.7303370786516854,
|
1924 |
+
"grad_norm": 2.372293297792043,
|
1925 |
+
"learning_rate": 2.056542406879957e-06,
|
1926 |
+
"loss": 0.3818,
|
1927 |
+
"step": 1365
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.7330123060460139,
|
1931 |
+
"grad_norm": 2.090955405220056,
|
1932 |
+
"learning_rate": 2.018925608072118e-06,
|
1933 |
+
"loss": 0.3542,
|
1934 |
+
"step": 1370
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.7356875334403424,
|
1938 |
+
"grad_norm": 2.2442087175829193,
|
1939 |
+
"learning_rate": 1.981568799733979e-06,
|
1940 |
+
"loss": 0.3657,
|
1941 |
+
"step": 1375
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.7383627608346709,
|
1945 |
+
"grad_norm": 2.2516703366046578,
|
1946 |
+
"learning_rate": 1.9444752398902874e-06,
|
1947 |
+
"loss": 0.3615,
|
1948 |
+
"step": 1380
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.7410379882289995,
|
1952 |
+
"grad_norm": 2.259664285200144,
|
1953 |
+
"learning_rate": 1.907648163606925e-06,
|
1954 |
+
"loss": 0.3587,
|
1955 |
+
"step": 1385
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.7437132156233279,
|
1959 |
+
"grad_norm": 2.4360594395654473,
|
1960 |
+
"learning_rate": 1.871090782708756e-06,
|
1961 |
+
"loss": 0.3603,
|
1962 |
+
"step": 1390
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.7463884430176565,
|
1966 |
+
"grad_norm": 2.247000162892966,
|
1967 |
+
"learning_rate": 1.834806285499519e-06,
|
1968 |
+
"loss": 0.3542,
|
1969 |
+
"step": 1395
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.7490636704119851,
|
1973 |
+
"grad_norm": 2.0975932964206425,
|
1974 |
+
"learning_rate": 1.7987978364837649e-06,
|
1975 |
+
"loss": 0.3557,
|
1976 |
+
"step": 1400
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.7517388978063135,
|
1980 |
+
"grad_norm": 2.104100339362327,
|
1981 |
+
"learning_rate": 1.7630685760908623e-06,
|
1982 |
+
"loss": 0.358,
|
1983 |
+
"step": 1405
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.7544141252006421,
|
1987 |
+
"grad_norm": 2.2966685174118324,
|
1988 |
+
"learning_rate": 1.727621620401112e-06,
|
1989 |
+
"loss": 0.3672,
|
1990 |
+
"step": 1410
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.7570893525949706,
|
1994 |
+
"grad_norm": 2.2434262781603422,
|
1995 |
+
"learning_rate": 1.6924600608739843e-06,
|
1996 |
+
"loss": 0.3632,
|
1997 |
+
"step": 1415
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.7597645799892991,
|
2001 |
+
"grad_norm": 2.3695918803743052,
|
2002 |
+
"learning_rate": 1.6575869640784998e-06,
|
2003 |
+
"loss": 0.3674,
|
2004 |
+
"step": 1420
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.7624398073836276,
|
2008 |
+
"grad_norm": 2.2896558172173096,
|
2009 |
+
"learning_rate": 1.6230053714257821e-06,
|
2010 |
+
"loss": 0.3615,
|
2011 |
+
"step": 1425
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.7651150347779562,
|
2015 |
+
"grad_norm": 2.132919841878146,
|
2016 |
+
"learning_rate": 1.588718298903803e-06,
|
2017 |
+
"loss": 0.3564,
|
2018 |
+
"step": 1430
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.7677902621722846,
|
2022 |
+
"grad_norm": 2.179194206405239,
|
2023 |
+
"learning_rate": 1.554728736814356e-06,
|
2024 |
+
"loss": 0.352,
|
2025 |
+
"step": 1435
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.7704654895666132,
|
2029 |
+
"grad_norm": 2.2259000819781063,
|
2030 |
+
"learning_rate": 1.5210396495122481e-06,
|
2031 |
+
"loss": 0.3505,
|
2032 |
+
"step": 1440
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.7731407169609417,
|
2036 |
+
"grad_norm": 2.1237345266350847,
|
2037 |
+
"learning_rate": 1.4876539751467806e-06,
|
2038 |
+
"loss": 0.3521,
|
2039 |
+
"step": 1445
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.7758159443552702,
|
2043 |
+
"grad_norm": 2.2438319275859304,
|
2044 |
+
"learning_rate": 1.45457462540549e-06,
|
2045 |
+
"loss": 0.3467,
|
2046 |
+
"step": 1450
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.7784911717495987,
|
2050 |
+
"grad_norm": 2.1311046073289326,
|
2051 |
+
"learning_rate": 1.4218044852602176e-06,
|
2052 |
+
"loss": 0.3517,
|
2053 |
+
"step": 1455
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.7811663991439273,
|
2057 |
+
"grad_norm": 2.151731010721367,
|
2058 |
+
"learning_rate": 1.3893464127154976e-06,
|
2059 |
+
"loss": 0.36,
|
2060 |
+
"step": 1460
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.7838416265382557,
|
2064 |
+
"grad_norm": 2.1371565977438203,
|
2065 |
+
"learning_rate": 1.3572032385592999e-06,
|
2066 |
+
"loss": 0.3466,
|
2067 |
+
"step": 1465
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.7865168539325843,
|
2071 |
+
"grad_norm": 2.1766415440270532,
|
2072 |
+
"learning_rate": 1.325377766116146e-06,
|
2073 |
+
"loss": 0.3488,
|
2074 |
+
"step": 1470
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.7891920813269128,
|
2078 |
+
"grad_norm": 2.02619887138196,
|
2079 |
+
"learning_rate": 1.293872771002625e-06,
|
2080 |
+
"loss": 0.3486,
|
2081 |
+
"step": 1475
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.7918673087212413,
|
2085 |
+
"grad_norm": 2.3827191729249737,
|
2086 |
+
"learning_rate": 1.2626910008853154e-06,
|
2087 |
+
"loss": 0.3703,
|
2088 |
+
"step": 1480
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.7945425361155698,
|
2092 |
+
"grad_norm": 2.2615580929872117,
|
2093 |
+
"learning_rate": 1.231835175241155e-06,
|
2094 |
+
"loss": 0.3562,
|
2095 |
+
"step": 1485
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.7972177635098984,
|
2099 |
+
"grad_norm": 2.229708858675527,
|
2100 |
+
"learning_rate": 1.2013079851202642e-06,
|
2101 |
+
"loss": 0.3624,
|
2102 |
+
"step": 1490
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.7998929909042268,
|
2106 |
+
"grad_norm": 2.333874491979822,
|
2107 |
+
"learning_rate": 1.1711120929112507e-06,
|
2108 |
+
"loss": 0.3404,
|
2109 |
+
"step": 1495
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.8025682182985554,
|
2113 |
+
"grad_norm": 1.947457938033558,
|
2114 |
+
"learning_rate": 1.141250132109009e-06,
|
2115 |
+
"loss": 0.3535,
|
2116 |
+
"step": 1500
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.8052434456928839,
|
2120 |
+
"grad_norm": 2.0935289084516615,
|
2121 |
+
"learning_rate": 1.1117247070850534e-06,
|
2122 |
+
"loss": 0.3406,
|
2123 |
+
"step": 1505
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.8079186730872124,
|
2127 |
+
"grad_norm": 2.2325175076090065,
|
2128 |
+
"learning_rate": 1.0825383928603656e-06,
|
2129 |
+
"loss": 0.3526,
|
2130 |
+
"step": 1510
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.8105939004815409,
|
2134 |
+
"grad_norm": 2.0095290479997647,
|
2135 |
+
"learning_rate": 1.0536937348808341e-06,
|
2136 |
+
"loss": 0.3409,
|
2137 |
+
"step": 1515
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.8132691278758695,
|
2141 |
+
"grad_norm": 1.9836588566813402,
|
2142 |
+
"learning_rate": 1.0251932487952437e-06,
|
2143 |
+
"loss": 0.339,
|
2144 |
+
"step": 1520
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.8159443552701979,
|
2148 |
+
"grad_norm": 2.26784995989598,
|
2149 |
+
"learning_rate": 9.97039420235884e-07,
|
2150 |
+
"loss": 0.3471,
|
2151 |
+
"step": 1525
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.8186195826645265,
|
2155 |
+
"grad_norm": 2.1193196211762846,
|
2156 |
+
"learning_rate": 9.692347046017647e-07,
|
2157 |
+
"loss": 0.344,
|
2158 |
+
"step": 1530
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.821294810058855,
|
2162 |
+
"grad_norm": 2.211413032254883,
|
2163 |
+
"learning_rate": 9.417815268444719e-07,
|
2164 |
+
"loss": 0.3441,
|
2165 |
+
"step": 1535
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.8239700374531835,
|
2169 |
+
"grad_norm": 1.9471769083672645,
|
2170 |
+
"learning_rate": 9.146822812566819e-07,
|
2171 |
+
"loss": 0.3502,
|
2172 |
+
"step": 1540
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.826645264847512,
|
2176 |
+
"grad_norm": 2.1918923218693105,
|
2177 |
+
"learning_rate": 8.879393312633405e-07,
|
2178 |
+
"loss": 0.3451,
|
2179 |
+
"step": 1545
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.8293204922418406,
|
2183 |
+
"grad_norm": 2.1371349511781506,
|
2184 |
+
"learning_rate": 8.615550092155478e-07,
|
2185 |
+
"loss": 0.3435,
|
2186 |
+
"step": 1550
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.831995719636169,
|
2190 |
+
"grad_norm": 2.0845002217133923,
|
2191 |
+
"learning_rate": 8.355316161871369e-07,
|
2192 |
+
"loss": 0.3466,
|
2193 |
+
"step": 1555
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.8346709470304976,
|
2197 |
+
"grad_norm": 2.25020629433754,
|
2198 |
+
"learning_rate": 8.098714217739928e-07,
|
2199 |
+
"loss": 0.3419,
|
2200 |
+
"step": 1560
|
2201 |
+
},
|
2202 |
+
{
|
2203 |
+
"epoch": 0.8373461744248261,
|
2204 |
+
"grad_norm": 2.0744989514612127,
|
2205 |
+
"learning_rate": 7.845766638961172e-07,
|
2206 |
+
"loss": 0.3473,
|
2207 |
+
"step": 1565
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 0.8400214018191546,
|
2211 |
+
"grad_norm": 3.639837300779382,
|
2212 |
+
"learning_rate": 7.596495486024402e-07,
|
2213 |
+
"loss": 0.3515,
|
2214 |
+
"step": 1570
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 0.8426966292134831,
|
2218 |
+
"grad_norm": 2.0191662732122104,
|
2219 |
+
"learning_rate": 7.350922498784335e-07,
|
2220 |
+
"loss": 0.3519,
|
2221 |
+
"step": 1575
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 0.8453718566078117,
|
2225 |
+
"grad_norm": 2.228031235416842,
|
2226 |
+
"learning_rate": 7.109069094565024e-07,
|
2227 |
+
"loss": 0.3448,
|
2228 |
+
"step": 1580
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 0.8480470840021401,
|
2232 |
+
"grad_norm": 2.523017960304099,
|
2233 |
+
"learning_rate": 6.870956366291998e-07,
|
2234 |
+
"loss": 0.3388,
|
2235 |
+
"step": 1585
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 0.8507223113964687,
|
2239 |
+
"grad_norm": 2.060587000595492,
|
2240 |
+
"learning_rate": 6.636605080652686e-07,
|
2241 |
+
"loss": 0.3487,
|
2242 |
+
"step": 1590
|
2243 |
+
},
|
2244 |
+
{
|
2245 |
+
"epoch": 0.8533975387907973,
|
2246 |
+
"grad_norm": 1.9701444425040686,
|
2247 |
+
"learning_rate": 6.406035676285244e-07,
|
2248 |
+
"loss": 0.3301,
|
2249 |
+
"step": 1595
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 0.8560727661851257,
|
2253 |
+
"grad_norm": 2.0633572735895873,
|
2254 |
+
"learning_rate": 6.179268261996052e-07,
|
2255 |
+
"loss": 0.3367,
|
2256 |
+
"step": 1600
|
2257 |
+
},
|
2258 |
+
{
|
2259 |
+
"epoch": 0.8587479935794543,
|
2260 |
+
"grad_norm": 2.1729206325004204,
|
2261 |
+
"learning_rate": 5.956322615005928e-07,
|
2262 |
+
"loss": 0.3463,
|
2263 |
+
"step": 1605
|
2264 |
+
},
|
2265 |
+
{
|
2266 |
+
"epoch": 0.8614232209737828,
|
2267 |
+
"grad_norm": 2.1697102226952585,
|
2268 |
+
"learning_rate": 5.737218179225318e-07,
|
2269 |
+
"loss": 0.3427,
|
2270 |
+
"step": 1610
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 0.8640984483681113,
|
2274 |
+
"grad_norm": 2.2114503479333942,
|
2275 |
+
"learning_rate": 5.521974063558477e-07,
|
2276 |
+
"loss": 0.3452,
|
2277 |
+
"step": 1615
|
2278 |
+
},
|
2279 |
+
{
|
2280 |
+
"epoch": 0.8667736757624398,
|
2281 |
+
"grad_norm": 2.0058575813876645,
|
2282 |
+
"learning_rate": 5.310609040236963e-07,
|
2283 |
+
"loss": 0.3471,
|
2284 |
+
"step": 1620
|
2285 |
+
},
|
2286 |
+
{
|
2287 |
+
"epoch": 0.8694489031567684,
|
2288 |
+
"grad_norm": 1.978522418614209,
|
2289 |
+
"learning_rate": 5.103141543182389e-07,
|
2290 |
+
"loss": 0.333,
|
2291 |
+
"step": 1625
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 0.8721241305510968,
|
2295 |
+
"grad_norm": 2.1706890891020127,
|
2296 |
+
"learning_rate": 4.89958966639878e-07,
|
2297 |
+
"loss": 0.3399,
|
2298 |
+
"step": 1630
|
2299 |
+
},
|
2300 |
+
{
|
2301 |
+
"epoch": 0.8747993579454254,
|
2302 |
+
"grad_norm": 2.102475974621354,
|
2303 |
+
"learning_rate": 4.6999711623944787e-07,
|
2304 |
+
"loss": 0.3295,
|
2305 |
+
"step": 1635
|
2306 |
+
},
|
2307 |
+
{
|
2308 |
+
"epoch": 0.8774745853397539,
|
2309 |
+
"grad_norm": 2.1350237073132243,
|
2310 |
+
"learning_rate": 4.504303440633928e-07,
|
2311 |
+
"loss": 0.3443,
|
2312 |
+
"step": 1640
|
2313 |
+
},
|
2314 |
+
{
|
2315 |
+
"epoch": 0.8801498127340824,
|
2316 |
+
"grad_norm": 2.0600510975030195,
|
2317 |
+
"learning_rate": 4.3126035660193076e-07,
|
2318 |
+
"loss": 0.3365,
|
2319 |
+
"step": 1645
|
2320 |
+
},
|
2321 |
+
{
|
2322 |
+
"epoch": 0.8828250401284109,
|
2323 |
+
"grad_norm": 2.236173654716049,
|
2324 |
+
"learning_rate": 4.124888257402243e-07,
|
2325 |
+
"loss": 0.3387,
|
2326 |
+
"step": 1650
|
2327 |
+
},
|
2328 |
+
{
|
2329 |
+
"epoch": 0.8855002675227395,
|
2330 |
+
"grad_norm": 1.996336345890085,
|
2331 |
+
"learning_rate": 3.9411738861256934e-07,
|
2332 |
+
"loss": 0.3403,
|
2333 |
+
"step": 1655
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 0.8881754949170679,
|
2337 |
+
"grad_norm": 1.9330066477895274,
|
2338 |
+
"learning_rate": 3.7614764745961377e-07,
|
2339 |
+
"loss": 0.3391,
|
2340 |
+
"step": 1660
|
2341 |
+
},
|
2342 |
+
{
|
2343 |
+
"epoch": 0.8908507223113965,
|
2344 |
+
"grad_norm": 2.1574712219106478,
|
2345 |
+
"learning_rate": 3.585811694886232e-07,
|
2346 |
+
"loss": 0.3337,
|
2347 |
+
"step": 1665
|
2348 |
+
},
|
2349 |
+
{
|
2350 |
+
"epoch": 0.893525949705725,
|
2351 |
+
"grad_norm": 2.1699077488308403,
|
2352 |
+
"learning_rate": 3.4141948673679593e-07,
|
2353 |
+
"loss": 0.3344,
|
2354 |
+
"step": 1670
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 0.8962011771000535,
|
2358 |
+
"grad_norm": 2.128864116785053,
|
2359 |
+
"learning_rate": 3.2466409593764734e-07,
|
2360 |
+
"loss": 0.3363,
|
2361 |
+
"step": 1675
|
2362 |
+
},
|
2363 |
+
{
|
2364 |
+
"epoch": 0.898876404494382,
|
2365 |
+
"grad_norm": 2.118472484473068,
|
2366 |
+
"learning_rate": 3.083164583904802e-07,
|
2367 |
+
"loss": 0.3415,
|
2368 |
+
"step": 1680
|
2369 |
+
},
|
2370 |
+
{
|
2371 |
+
"epoch": 0.9015516318887106,
|
2372 |
+
"grad_norm": 2.0200831474062886,
|
2373 |
+
"learning_rate": 2.923779998329318e-07,
|
2374 |
+
"loss": 0.3224,
|
2375 |
+
"step": 1685
|
2376 |
+
},
|
2377 |
+
{
|
2378 |
+
"epoch": 0.904226859283039,
|
2379 |
+
"grad_norm": 2.144747824231417,
|
2380 |
+
"learning_rate": 2.76850110316636e-07,
|
2381 |
+
"loss": 0.3343,
|
2382 |
+
"step": 1690
|
2383 |
+
},
|
2384 |
+
{
|
2385 |
+
"epoch": 0.9069020866773676,
|
2386 |
+
"grad_norm": 2.122020641145892,
|
2387 |
+
"learning_rate": 2.617341440859883e-07,
|
2388 |
+
"loss": 0.3465,
|
2389 |
+
"step": 1695
|
2390 |
+
},
|
2391 |
+
{
|
2392 |
+
"epoch": 0.9095773140716961,
|
2393 |
+
"grad_norm": 2.072786373493989,
|
2394 |
+
"learning_rate": 2.470314194600376e-07,
|
2395 |
+
"loss": 0.3231,
|
2396 |
+
"step": 1700
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 0.9122525414660246,
|
2400 |
+
"grad_norm": 1.9359758916982563,
|
2401 |
+
"learning_rate": 2.3274321871751436e-07,
|
2402 |
+
"loss": 0.3353,
|
2403 |
+
"step": 1705
|
2404 |
+
},
|
2405 |
+
{
|
2406 |
+
"epoch": 0.9149277688603531,
|
2407 |
+
"grad_norm": 1.9922992143020988,
|
2408 |
+
"learning_rate": 2.1887078798499272e-07,
|
2409 |
+
"loss": 0.3413,
|
2410 |
+
"step": 1710
|
2411 |
+
},
|
2412 |
+
{
|
2413 |
+
"epoch": 0.9176029962546817,
|
2414 |
+
"grad_norm": 2.334136482586755,
|
2415 |
+
"learning_rate": 2.0541533712821527e-07,
|
2416 |
+
"loss": 0.3354,
|
2417 |
+
"step": 1715
|
2418 |
+
},
|
2419 |
+
{
|
2420 |
+
"epoch": 0.9202782236490101,
|
2421 |
+
"grad_norm": 2.2163195568166145,
|
2422 |
+
"learning_rate": 1.923780396465741e-07,
|
2423 |
+
"loss": 0.3412,
|
2424 |
+
"step": 1720
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 0.9229534510433387,
|
2428 |
+
"grad_norm": 1.9231313598051207,
|
2429 |
+
"learning_rate": 1.7976003257076823e-07,
|
2430 |
+
"loss": 0.3385,
|
2431 |
+
"step": 1725
|
2432 |
+
},
|
2433 |
+
{
|
2434 |
+
"epoch": 0.9256286784376672,
|
2435 |
+
"grad_norm": 2.0705877113088844,
|
2436 |
+
"learning_rate": 1.6756241636363413e-07,
|
2437 |
+
"loss": 0.3298,
|
2438 |
+
"step": 1730
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 0.9283039058319957,
|
2442 |
+
"grad_norm": 2.057376654150935,
|
2443 |
+
"learning_rate": 1.557862548241762e-07,
|
2444 |
+
"loss": 0.3327,
|
2445 |
+
"step": 1735
|
2446 |
+
},
|
2447 |
+
{
|
2448 |
+
"epoch": 0.9309791332263242,
|
2449 |
+
"grad_norm": 2.107808663106814,
|
2450 |
+
"learning_rate": 1.4443257499478447e-07,
|
2451 |
+
"loss": 0.3416,
|
2452 |
+
"step": 1740
|
2453 |
+
},
|
2454 |
+
{
|
2455 |
+
"epoch": 0.9336543606206528,
|
2456 |
+
"grad_norm": 1.9829198272876933,
|
2457 |
+
"learning_rate": 1.3350236707166508e-07,
|
2458 |
+
"loss": 0.3414,
|
2459 |
+
"step": 1745
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 0.9363295880149812,
|
2463 |
+
"grad_norm": 2.056040338556245,
|
2464 |
+
"learning_rate": 1.229965843184805e-07,
|
2465 |
+
"loss": 0.3413,
|
2466 |
+
"step": 1750
|
2467 |
+
},
|
2468 |
+
{
|
2469 |
+
"epoch": 0.9390048154093098,
|
2470 |
+
"grad_norm": 2.337178376808903,
|
2471 |
+
"learning_rate": 1.1291614298321097e-07,
|
2472 |
+
"loss": 0.3409,
|
2473 |
+
"step": 1755
|
2474 |
+
},
|
2475 |
+
{
|
2476 |
+
"epoch": 0.9416800428036383,
|
2477 |
+
"grad_norm": 2.1520247260228746,
|
2478 |
+
"learning_rate": 1.0326192221824738e-07,
|
2479 |
+
"loss": 0.3232,
|
2480 |
+
"step": 1760
|
2481 |
+
},
|
2482 |
+
{
|
2483 |
+
"epoch": 0.9443552701979668,
|
2484 |
+
"grad_norm": 2.142605939615871,
|
2485 |
+
"learning_rate": 9.403476400371425e-08,
|
2486 |
+
"loss": 0.3397,
|
2487 |
+
"step": 1765
|
2488 |
+
},
|
2489 |
+
{
|
2490 |
+
"epoch": 0.9470304975922953,
|
2491 |
+
"grad_norm": 2.0450418837244366,
|
2492 |
+
"learning_rate": 8.523547307404179e-08,
|
2493 |
+
"loss": 0.3368,
|
2494 |
+
"step": 1770
|
2495 |
+
},
|
2496 |
+
{
|
2497 |
+
"epoch": 0.9497057249866239,
|
2498 |
+
"grad_norm": 2.0614174232017923,
|
2499 |
+
"learning_rate": 7.686481684777758e-08,
|
2500 |
+
"loss": 0.3321,
|
2501 |
+
"step": 1775
|
2502 |
+
},
|
2503 |
+
{
|
2504 |
+
"epoch": 0.9523809523809523,
|
2505 |
+
"grad_norm": 2.1361641337619015,
|
2506 |
+
"learning_rate": 6.89235253606596e-08,
|
2507 |
+
"loss": 0.3154,
|
2508 |
+
"step": 1780
|
2509 |
+
},
|
2510 |
+
{
|
2511 |
+
"epoch": 0.9550561797752809,
|
2512 |
+
"grad_norm": 2.0514520677857244,
|
2513 |
+
"learning_rate": 6.141229120194714e-08,
|
2514 |
+
"loss": 0.3386,
|
2515 |
+
"step": 1785
|
2516 |
+
},
|
2517 |
+
{
|
2518 |
+
"epoch": 0.9577314071696095,
|
2519 |
+
"grad_norm": 2.069777118733485,
|
2520 |
+
"learning_rate": 5.4331769454016306e-08,
|
2521 |
+
"loss": 0.3319,
|
2522 |
+
"step": 1790
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 0.9604066345639379,
|
2526 |
+
"grad_norm": 2.136465423921526,
|
2527 |
+
"learning_rate": 4.76825776352291e-08,
|
2528 |
+
"loss": 0.338,
|
2529 |
+
"step": 1795
|
2530 |
+
},
|
2531 |
+
{
|
2532 |
+
"epoch": 0.9630818619582665,
|
2533 |
+
"grad_norm": 2.0522216771136934,
|
2534 |
+
"learning_rate": 4.1465295646076484e-08,
|
2535 |
+
"loss": 0.3298,
|
2536 |
+
"step": 1800
|
2537 |
+
},
|
2538 |
+
{
|
2539 |
+
"epoch": 0.965757089352595,
|
2540 |
+
"grad_norm": 2.1070855069769094,
|
2541 |
+
"learning_rate": 3.568046571860384e-08,
|
2542 |
+
"loss": 0.3291,
|
2543 |
+
"step": 1805
|
2544 |
+
},
|
2545 |
+
{
|
2546 |
+
"epoch": 0.9684323167469235,
|
2547 |
+
"grad_norm": 2.2197166876041576,
|
2548 |
+
"learning_rate": 3.0328592369120443e-08,
|
2549 |
+
"loss": 0.3373,
|
2550 |
+
"step": 1810
|
2551 |
+
},
|
2552 |
+
{
|
2553 |
+
"epoch": 0.971107544141252,
|
2554 |
+
"grad_norm": 1.8711860549385024,
|
2555 |
+
"learning_rate": 2.541014235419914e-08,
|
2556 |
+
"loss": 0.3314,
|
2557 |
+
"step": 1815
|
2558 |
+
},
|
2559 |
+
{
|
2560 |
+
"epoch": 0.9737827715355806,
|
2561 |
+
"grad_norm": 2.0398927890227805,
|
2562 |
+
"learning_rate": 2.0925544629967763e-08,
|
2563 |
+
"loss": 0.337,
|
2564 |
+
"step": 1820
|
2565 |
+
},
|
2566 |
+
{
|
2567 |
+
"epoch": 0.976457998929909,
|
2568 |
+
"grad_norm": 2.0934381794140253,
|
2569 |
+
"learning_rate": 1.6875190314700197e-08,
|
2570 |
+
"loss": 0.3348,
|
2571 |
+
"step": 1825
|
2572 |
+
},
|
2573 |
+
{
|
2574 |
+
"epoch": 0.9791332263242376,
|
2575 |
+
"grad_norm": 2.0969584920178517,
|
2576 |
+
"learning_rate": 1.3259432654703641e-08,
|
2577 |
+
"loss": 0.3338,
|
2578 |
+
"step": 1830
|
2579 |
+
},
|
2580 |
+
{
|
2581 |
+
"epoch": 0.9818084537185661,
|
2582 |
+
"grad_norm": 1.859307599341638,
|
2583 |
+
"learning_rate": 1.0078586993511052e-08,
|
2584 |
+
"loss": 0.3373,
|
2585 |
+
"step": 1835
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 0.9844836811128946,
|
2589 |
+
"grad_norm": 2.085094787953033,
|
2590 |
+
"learning_rate": 7.332930744380906e-09,
|
2591 |
+
"loss": 0.331,
|
2592 |
+
"step": 1840
|
2593 |
+
},
|
2594 |
+
{
|
2595 |
+
"epoch": 0.9871589085072231,
|
2596 |
+
"grad_norm": 2.0189271533665263,
|
2597 |
+
"learning_rate": 5.0227033660987804e-09,
|
2598 |
+
"loss": 0.3259,
|
2599 |
+
"step": 1845
|
2600 |
+
},
|
2601 |
+
{
|
2602 |
+
"epoch": 0.9898341359015517,
|
2603 |
+
"grad_norm": 2.087692299427238,
|
2604 |
+
"learning_rate": 3.1481063420985e-09,
|
2605 |
+
"loss": 0.3324,
|
2606 |
+
"step": 1850
|
2607 |
+
},
|
2608 |
+
{
|
2609 |
+
"epoch": 0.9925093632958801,
|
2610 |
+
"grad_norm": 2.1232755165842856,
|
2611 |
+
"learning_rate": 1.7093031628850899e-09,
|
2612 |
+
"loss": 0.328,
|
2613 |
+
"step": 1855
|
2614 |
+
},
|
2615 |
+
{
|
2616 |
+
"epoch": 0.9951845906902087,
|
2617 |
+
"grad_norm": 2.2098470869075864,
|
2618 |
+
"learning_rate": 7.064193117806151e-10,
|
2619 |
+
"loss": 0.3408,
|
2620 |
+
"step": 1860
|
2621 |
+
},
|
2622 |
+
{
|
2623 |
+
"epoch": 0.9978598180845372,
|
2624 |
+
"grad_norm": 2.177061948752331,
|
2625 |
+
"learning_rate": 1.3954225397516673e-10,
|
2626 |
+
"loss": 0.3345,
|
2627 |
+
"step": 1865
|
2628 |
+
},
|
2629 |
+
{
|
2630 |
+
"epoch": 1.0,
|
2631 |
+
"eval_runtime": 3.4105,
|
2632 |
+
"eval_samples_per_second": 2.932,
|
2633 |
+
"eval_steps_per_second": 0.88,
|
2634 |
+
"step": 1869
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 1.0,
|
2638 |
+
"step": 1869,
|
2639 |
+
"total_flos": 195665288232960.0,
|
2640 |
+
"train_loss": 0.4974420365322842,
|
2641 |
+
"train_runtime": 16839.356,
|
2642 |
+
"train_samples_per_second": 1.776,
|
2643 |
+
"train_steps_per_second": 0.111
|
2644 |
+
}
|
2645 |
+
],
|
2646 |
+
"logging_steps": 5,
|
2647 |
+
"max_steps": 1869,
|
2648 |
+
"num_input_tokens_seen": 0,
|
2649 |
+
"num_train_epochs": 1,
|
2650 |
+
"save_steps": 100,
|
2651 |
+
"stateful_callbacks": {
|
2652 |
+
"TrainerControl": {
|
2653 |
+
"args": {
|
2654 |
+
"should_epoch_stop": false,
|
2655 |
+
"should_evaluate": false,
|
2656 |
+
"should_log": false,
|
2657 |
+
"should_save": true,
|
2658 |
+
"should_training_stop": true
|
2659 |
+
},
|
2660 |
+
"attributes": {}
|
2661 |
+
}
|
2662 |
+
},
|
2663 |
+
"total_flos": 195665288232960.0,
|
2664 |
+
"train_batch_size": 4,
|
2665 |
+
"trial_name": null,
|
2666 |
+
"trial_params": null
|
2667 |
+
}
|