beamaia commited on
Commit
44c688f
1 Parent(s): 508f621

Training in progress, step 400, checkpoint

Browse files
checkpoint-400/README.md ADDED
@@ -0,0 +1,204 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: HuggingFaceH4/zephyr-7b-beta
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+
201
+
202
+ ### Framework versions
203
+
204
+ - PEFT 0.10.0
checkpoint-400/adapter_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "HuggingFaceH4/zephyr-7b-beta",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 32,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 16,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "o_proj",
24
+ "v_proj",
25
+ "k_proj",
26
+ "q_proj"
27
+ ],
28
+ "task_type": "CAUSAL_LM",
29
+ "use_dora": false,
30
+ "use_rslora": false
31
+ }
checkpoint-400/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d11257f5a31b8a7a52ea707da5307c9e48431db2bef060b3e1ca4dbd212bd891
3
+ size 54560368
checkpoint-400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0049846f140b51f51e3375071741124f8c7170d479a1968da978ee7f849a5011
3
+ size 109267450
checkpoint-400/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f3803bff3f596c03b55881de967a825b5734e4a581739164f9cb9e7fd1aee89
3
+ size 14512
checkpoint-400/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d768a04b798e2ca42effbe096b8e4481f32a402a9125a2ced390586dab8eb29e
3
+ size 14512
checkpoint-400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0036d82f2e21e3394011d2fd21cbbd9af467f37a026201c919dae62d508ef9ba
3
+ size 1064
checkpoint-400/special_tokens_map.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<unk>",
4
+ "<s>",
5
+ "</s>"
6
+ ],
7
+ "bos_token": {
8
+ "content": "<s>",
9
+ "lstrip": false,
10
+ "normalized": false,
11
+ "rstrip": false,
12
+ "single_word": false
13
+ },
14
+ "eos_token": {
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "pad_token": "<unk>",
22
+ "unk_token": {
23
+ "content": "<unk>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false
28
+ }
29
+ }
checkpoint-400/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-400/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
checkpoint-400/tokenizer_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [
31
+ "<unk>",
32
+ "<s>",
33
+ "</s>"
34
+ ],
35
+ "bos_token": "<s>",
36
+ "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
37
+ "clean_up_tokenization_spaces": false,
38
+ "eos_token": "</s>",
39
+ "legacy": true,
40
+ "max_lenght": 8192,
41
+ "model_max_length": 1000000000000000019884624838656,
42
+ "pad_token": "<unk>",
43
+ "padding": true,
44
+ "sp_model_kwargs": {},
45
+ "spaces_between_special_tokens": false,
46
+ "tokenizer_class": "LlamaTokenizer",
47
+ "truncation_side": "left",
48
+ "unk_token": "<unk>",
49
+ "use_default_system_prompt": true
50
+ }
checkpoint-400/trainer_state.json ADDED
@@ -0,0 +1,393 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.014748962596058846,
3
+ "best_model_checkpoint": "./Zephyr/27-03-24-Weni-kto-test_WeniGPT Experiment using KTO trainer with no collator-2_max_steps-786_batch_16_2024-03-27_ppid_1885/checkpoint-300",
4
+ "epoch": 3.0476190476190474,
5
+ "eval_steps": 50,
6
+ "global_step": 400,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.15,
13
+ "grad_norm": 2.2224462032318115,
14
+ "kl": 0.054589949548244476,
15
+ "learning_rate": 0.0001666666666666667,
16
+ "logps/chosen": NaN,
17
+ "logps/rejected": NaN,
18
+ "loss": 704.2891,
19
+ "rewards/chosen": NaN,
20
+ "rewards/margins": NaN,
21
+ "rewards/rejected": NaN,
22
+ "step": 20
23
+ },
24
+ {
25
+ "epoch": 0.3,
26
+ "grad_norm": 2.9781014919281006,
27
+ "kl": 0.014553820714354515,
28
+ "learning_rate": 0.00019580052493438322,
29
+ "logps/chosen": NaN,
30
+ "logps/rejected": NaN,
31
+ "loss": 371.0873,
32
+ "rewards/chosen": NaN,
33
+ "rewards/margins": NaN,
34
+ "rewards/rejected": NaN,
35
+ "step": 40
36
+ },
37
+ {
38
+ "epoch": 0.38,
39
+ "eval_kl": 0.014955863356590271,
40
+ "eval_logps/chosen": -139.6813507080078,
41
+ "eval_logps/rejected": -288.11480712890625,
42
+ "eval_loss": 0.0440419502556324,
43
+ "eval_rewards/chosen": 4.680752754211426,
44
+ "eval_rewards/margins": 14.199661254882812,
45
+ "eval_rewards/rejected": -9.518909454345703,
46
+ "eval_runtime": 437.1874,
47
+ "eval_samples_per_second": 1.07,
48
+ "eval_steps_per_second": 0.268,
49
+ "step": 50
50
+ },
51
+ {
52
+ "epoch": 0.46,
53
+ "grad_norm": 0.7002557516098022,
54
+ "kl": 0.36391472816467285,
55
+ "learning_rate": 0.0001905511811023622,
56
+ "logps/chosen": NaN,
57
+ "logps/rejected": NaN,
58
+ "loss": 165.1605,
59
+ "rewards/chosen": NaN,
60
+ "rewards/margins": NaN,
61
+ "rewards/rejected": NaN,
62
+ "step": 60
63
+ },
64
+ {
65
+ "epoch": 0.61,
66
+ "grad_norm": 0.5644651651382446,
67
+ "kl": 0.01063810009509325,
68
+ "learning_rate": 0.00018556430446194227,
69
+ "logps/chosen": NaN,
70
+ "logps/rejected": NaN,
71
+ "loss": 89.197,
72
+ "rewards/chosen": NaN,
73
+ "rewards/margins": NaN,
74
+ "rewards/rejected": NaN,
75
+ "step": 80
76
+ },
77
+ {
78
+ "epoch": 0.76,
79
+ "grad_norm": 0.18941974639892578,
80
+ "kl": 0.0,
81
+ "learning_rate": 0.0001805774278215223,
82
+ "logps/chosen": NaN,
83
+ "logps/rejected": NaN,
84
+ "loss": 57.9834,
85
+ "rewards/chosen": NaN,
86
+ "rewards/margins": NaN,
87
+ "rewards/rejected": NaN,
88
+ "step": 100
89
+ },
90
+ {
91
+ "epoch": 0.76,
92
+ "eval_kl": 0.0,
93
+ "eval_logps/chosen": -135.0946807861328,
94
+ "eval_logps/rejected": -511.870361328125,
95
+ "eval_loss": 0.027495555579662323,
96
+ "eval_rewards/chosen": 5.139419078826904,
97
+ "eval_rewards/margins": 37.033878326416016,
98
+ "eval_rewards/rejected": -31.89446258544922,
99
+ "eval_runtime": 437.2677,
100
+ "eval_samples_per_second": 1.07,
101
+ "eval_steps_per_second": 0.268,
102
+ "step": 100
103
+ },
104
+ {
105
+ "epoch": 0.91,
106
+ "grad_norm": 1.2737089395523071,
107
+ "kl": 0.0,
108
+ "learning_rate": 0.00017532808398950132,
109
+ "logps/chosen": NaN,
110
+ "logps/rejected": NaN,
111
+ "loss": 1101.4947,
112
+ "rewards/chosen": NaN,
113
+ "rewards/margins": NaN,
114
+ "rewards/rejected": NaN,
115
+ "step": 120
116
+ },
117
+ {
118
+ "epoch": 1.07,
119
+ "grad_norm": 0.10652273148298264,
120
+ "kl": 0.0,
121
+ "learning_rate": 0.00017007874015748033,
122
+ "logps/chosen": NaN,
123
+ "logps/rejected": NaN,
124
+ "loss": 37.3685,
125
+ "rewards/chosen": NaN,
126
+ "rewards/margins": NaN,
127
+ "rewards/rejected": NaN,
128
+ "step": 140
129
+ },
130
+ {
131
+ "epoch": 1.14,
132
+ "eval_kl": 0.0,
133
+ "eval_logps/chosen": -133.9324951171875,
134
+ "eval_logps/rejected": -464.85992431640625,
135
+ "eval_loss": 0.019584182649850845,
136
+ "eval_rewards/chosen": 5.25563907623291,
137
+ "eval_rewards/margins": 32.449058532714844,
138
+ "eval_rewards/rejected": -27.19342041015625,
139
+ "eval_runtime": 437.0658,
140
+ "eval_samples_per_second": 1.071,
141
+ "eval_steps_per_second": 0.268,
142
+ "step": 150
143
+ },
144
+ {
145
+ "epoch": 1.22,
146
+ "grad_norm": 0.060089632868766785,
147
+ "kl": 0.21830137073993683,
148
+ "learning_rate": 0.00016482939632545934,
149
+ "logps/chosen": NaN,
150
+ "logps/rejected": NaN,
151
+ "loss": 91.1641,
152
+ "rewards/chosen": NaN,
153
+ "rewards/margins": NaN,
154
+ "rewards/rejected": NaN,
155
+ "step": 160
156
+ },
157
+ {
158
+ "epoch": 1.37,
159
+ "grad_norm": 0.08910108357667923,
160
+ "kl": 0.03056584671139717,
161
+ "learning_rate": 0.00015958005249343832,
162
+ "logps/chosen": NaN,
163
+ "logps/rejected": NaN,
164
+ "loss": 8.2078,
165
+ "rewards/chosen": NaN,
166
+ "rewards/margins": NaN,
167
+ "rewards/rejected": NaN,
168
+ "step": 180
169
+ },
170
+ {
171
+ "epoch": 1.52,
172
+ "grad_norm": 0.025327660143375397,
173
+ "kl": 0.0,
174
+ "learning_rate": 0.00015433070866141733,
175
+ "logps/chosen": NaN,
176
+ "logps/rejected": NaN,
177
+ "loss": 3.6561,
178
+ "rewards/chosen": NaN,
179
+ "rewards/margins": NaN,
180
+ "rewards/rejected": NaN,
181
+ "step": 200
182
+ },
183
+ {
184
+ "epoch": 1.52,
185
+ "eval_kl": 0.0,
186
+ "eval_logps/chosen": -132.1833038330078,
187
+ "eval_logps/rejected": -419.2353820800781,
188
+ "eval_loss": 0.016207611188292503,
189
+ "eval_rewards/chosen": 5.430557727813721,
190
+ "eval_rewards/margins": 28.061521530151367,
191
+ "eval_rewards/rejected": -22.630962371826172,
192
+ "eval_runtime": 437.1734,
193
+ "eval_samples_per_second": 1.071,
194
+ "eval_steps_per_second": 0.268,
195
+ "step": 200
196
+ },
197
+ {
198
+ "epoch": 1.68,
199
+ "grad_norm": 0.04120240360498428,
200
+ "kl": 0.04972388595342636,
201
+ "learning_rate": 0.00014908136482939634,
202
+ "logps/chosen": NaN,
203
+ "logps/rejected": NaN,
204
+ "loss": 21.787,
205
+ "rewards/chosen": NaN,
206
+ "rewards/margins": NaN,
207
+ "rewards/rejected": NaN,
208
+ "step": 220
209
+ },
210
+ {
211
+ "epoch": 1.83,
212
+ "grad_norm": 0.043324064463377,
213
+ "kl": 0.09086960554122925,
214
+ "learning_rate": 0.00014383202099737535,
215
+ "logps/chosen": NaN,
216
+ "logps/rejected": NaN,
217
+ "loss": 59.5367,
218
+ "rewards/chosen": NaN,
219
+ "rewards/margins": NaN,
220
+ "rewards/rejected": NaN,
221
+ "step": 240
222
+ },
223
+ {
224
+ "epoch": 1.9,
225
+ "eval_kl": 0.0,
226
+ "eval_logps/chosen": -129.1339111328125,
227
+ "eval_logps/rejected": -504.5447998046875,
228
+ "eval_loss": 0.01432154793292284,
229
+ "eval_rewards/chosen": 5.7354960441589355,
230
+ "eval_rewards/margins": 36.89740753173828,
231
+ "eval_rewards/rejected": -31.161909103393555,
232
+ "eval_runtime": 437.5542,
233
+ "eval_samples_per_second": 1.07,
234
+ "eval_steps_per_second": 0.267,
235
+ "step": 250
236
+ },
237
+ {
238
+ "epoch": 1.98,
239
+ "grad_norm": 0.028953969478607178,
240
+ "kl": 0.0,
241
+ "learning_rate": 0.00013858267716535433,
242
+ "logps/chosen": NaN,
243
+ "logps/rejected": NaN,
244
+ "loss": 5.7556,
245
+ "rewards/chosen": NaN,
246
+ "rewards/margins": NaN,
247
+ "rewards/rejected": NaN,
248
+ "step": 260
249
+ },
250
+ {
251
+ "epoch": 2.13,
252
+ "grad_norm": 0.022753821685910225,
253
+ "kl": 0.18936340510845184,
254
+ "learning_rate": 0.00013333333333333334,
255
+ "logps/chosen": NaN,
256
+ "logps/rejected": NaN,
257
+ "loss": 4.7774,
258
+ "rewards/chosen": NaN,
259
+ "rewards/margins": NaN,
260
+ "rewards/rejected": NaN,
261
+ "step": 280
262
+ },
263
+ {
264
+ "epoch": 2.29,
265
+ "grad_norm": 0.038736581802368164,
266
+ "kl": 0.0,
267
+ "learning_rate": 0.00012808398950131235,
268
+ "logps/chosen": NaN,
269
+ "logps/rejected": NaN,
270
+ "loss": 13.1891,
271
+ "rewards/chosen": NaN,
272
+ "rewards/margins": NaN,
273
+ "rewards/rejected": NaN,
274
+ "step": 300
275
+ },
276
+ {
277
+ "epoch": 2.29,
278
+ "eval_kl": 0.0,
279
+ "eval_logps/chosen": -130.3461151123047,
280
+ "eval_logps/rejected": -503.4655456542969,
281
+ "eval_loss": 0.014748962596058846,
282
+ "eval_rewards/chosen": 5.614275932312012,
283
+ "eval_rewards/margins": 36.66826248168945,
284
+ "eval_rewards/rejected": -31.05398178100586,
285
+ "eval_runtime": 437.6283,
286
+ "eval_samples_per_second": 1.069,
287
+ "eval_steps_per_second": 0.267,
288
+ "step": 300
289
+ },
290
+ {
291
+ "epoch": 2.44,
292
+ "grad_norm": 0.015456304885447025,
293
+ "kl": 0.016447115689516068,
294
+ "learning_rate": 0.00012283464566929136,
295
+ "logps/chosen": NaN,
296
+ "logps/rejected": NaN,
297
+ "loss": 2.946,
298
+ "rewards/chosen": NaN,
299
+ "rewards/margins": NaN,
300
+ "rewards/rejected": NaN,
301
+ "step": 320
302
+ },
303
+ {
304
+ "epoch": 2.59,
305
+ "grad_norm": 0.03844680264592171,
306
+ "kl": 0.0,
307
+ "learning_rate": 0.00011758530183727034,
308
+ "logps/chosen": NaN,
309
+ "logps/rejected": NaN,
310
+ "loss": 3.8532,
311
+ "rewards/chosen": NaN,
312
+ "rewards/margins": NaN,
313
+ "rewards/rejected": NaN,
314
+ "step": 340
315
+ },
316
+ {
317
+ "epoch": 2.67,
318
+ "eval_kl": 0.0,
319
+ "eval_logps/chosen": -127.6288833618164,
320
+ "eval_logps/rejected": -457.08013916015625,
321
+ "eval_loss": 0.013063323684036732,
322
+ "eval_rewards/chosen": 5.886001110076904,
323
+ "eval_rewards/margins": 32.30144119262695,
324
+ "eval_rewards/rejected": -26.415441513061523,
325
+ "eval_runtime": 437.2,
326
+ "eval_samples_per_second": 1.07,
327
+ "eval_steps_per_second": 0.268,
328
+ "step": 350
329
+ },
330
+ {
331
+ "epoch": 2.74,
332
+ "grad_norm": 0.013407759368419647,
333
+ "kl": 0.0,
334
+ "learning_rate": 0.00011233595800524934,
335
+ "logps/chosen": NaN,
336
+ "logps/rejected": NaN,
337
+ "loss": 14.6044,
338
+ "rewards/chosen": NaN,
339
+ "rewards/margins": NaN,
340
+ "rewards/rejected": NaN,
341
+ "step": 360
342
+ },
343
+ {
344
+ "epoch": 2.9,
345
+ "grad_norm": 0.1375456601381302,
346
+ "kl": 0.042511165142059326,
347
+ "learning_rate": 0.00010708661417322836,
348
+ "logps/chosen": NaN,
349
+ "logps/rejected": NaN,
350
+ "loss": 20.8824,
351
+ "rewards/chosen": NaN,
352
+ "rewards/margins": NaN,
353
+ "rewards/rejected": NaN,
354
+ "step": 380
355
+ },
356
+ {
357
+ "epoch": 3.05,
358
+ "grad_norm": 0.012805495411157608,
359
+ "kl": 0.0,
360
+ "learning_rate": 0.00010183727034120735,
361
+ "logps/chosen": NaN,
362
+ "logps/rejected": NaN,
363
+ "loss": 3.7678,
364
+ "rewards/chosen": NaN,
365
+ "rewards/margins": NaN,
366
+ "rewards/rejected": NaN,
367
+ "step": 400
368
+ },
369
+ {
370
+ "epoch": 3.05,
371
+ "eval_kl": 0.0,
372
+ "eval_logps/chosen": -127.17108917236328,
373
+ "eval_logps/rejected": -460.4493408203125,
374
+ "eval_loss": 0.016150476410984993,
375
+ "eval_rewards/chosen": 5.931778430938721,
376
+ "eval_rewards/margins": 32.684139251708984,
377
+ "eval_rewards/rejected": -26.75235939025879,
378
+ "eval_runtime": 437.1823,
379
+ "eval_samples_per_second": 1.07,
380
+ "eval_steps_per_second": 0.268,
381
+ "step": 400
382
+ }
383
+ ],
384
+ "logging_steps": 20,
385
+ "max_steps": 786,
386
+ "num_input_tokens_seen": 0,
387
+ "num_train_epochs": 6,
388
+ "save_steps": 100,
389
+ "total_flos": 0.0,
390
+ "train_batch_size": 2,
391
+ "trial_name": null,
392
+ "trial_params": null
393
+ }
checkpoint-400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c04380dac1b289b8afea7297ea9ea9e1ac98da4787c8cec5f2868e7236a8fefb
3
+ size 5560