chchen commited on
Commit
e29bf33
·
verified ·
1 Parent(s): 32da209

Upload 13 files

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mistralai/Ministral-8B-Instruct-2410
3
+ library_name: peft
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
+ ### Framework versions
201
+
202
+ - PEFT 0.12.0
adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "mistralai/Ministral-8B-Instruct-2410",
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": 16,
14
+ "lora_dropout": 0.0,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 8,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "gate_proj",
24
+ "down_proj",
25
+ "v_proj",
26
+ "k_proj",
27
+ "up_proj",
28
+ "q_proj",
29
+ "o_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:071c31423afacdb75caf99b80b90d669af52a35a2dd07a16cb7cc423ea0d93f6
3
+ size 87360584
llama3_lora_ppo.yaml ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ### model
2
+ model_name_or_path: mistralai/Ministral-8B-Instruct-2410
3
+ reward_model: chchen/Ministral-8B-Instruct-2410-reward-1000
4
+ trust_remote_code: true
5
+
6
+ ### method
7
+ stage: ppo
8
+ do_train: true
9
+ finetuning_type: lora
10
+ lora_target: all
11
+
12
+ ### dataset
13
+ dataset: bct_non_cot_sft_1000
14
+ dataset_dir: data_private
15
+ template: mistral
16
+ cutoff_len: 1024
17
+ # max_samples: 1000
18
+ overwrite_cache: true
19
+ preprocessing_num_workers: 16
20
+
21
+ ### output
22
+ output_dir: saves/Ministral-8B-Instruct-2410/ppo-1000/train
23
+ logging_steps: 10
24
+ save_steps: 50
25
+ plot_loss: true
26
+ overwrite_output_dir: true
27
+ save_total_limit: 3
28
+ push_to_hub: true
29
+ hub_model_id: chchen/Ministral-8B-Instruct-2410-ppo-1000
30
+
31
+ ### train
32
+ per_device_train_batch_size: 4
33
+ gradient_accumulation_steps: 8
34
+ learning_rate: 1.0e-5
35
+ num_train_epochs: 10.0
36
+ lr_scheduler_type: cosine
37
+ warmup_ratio: 0.1
38
+ bf16: true
39
+ ddp_timeout: 180000000
40
+
41
+ ### generate
42
+ max_new_tokens: 512
43
+ top_k: 0
44
+ top_p: 0.9
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7edbeaf20dd7f571b5dd1c54d9ace4f9b6299127cc7ba2afb14a6d51a4a79a4
3
+ size 17078136
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
trainer_log.jsonl ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 20, "total_steps": 310, "loss": 4.6766, "reward": 8.9945, "lr": 5.9375e-06, "epoch": 0.61, "percentage": 6.45, "elapsed_time": "0:38:37", "remaining_time": "9:20:07"}
2
+ {"current_steps": 30, "total_steps": 310, "loss": 4.0729, "reward": 9.3216, "lr": 9.0625e-06, "epoch": 0.94, "percentage": 9.68, "elapsed_time": "0:54:25", "remaining_time": "8:27:55"}
3
+ {"current_steps": 40, "total_steps": 310, "loss": 3.0501, "reward": 9.949, "lr": 9.985430661522333e-06, "epoch": 1.26, "percentage": 12.9, "elapsed_time": "1:10:44", "remaining_time": "7:57:28"}
4
+ {"current_steps": 50, "total_steps": 310, "loss": 1.148, "reward": 11.1044, "lr": 9.914274958326507e-06, "epoch": 1.58, "percentage": 16.13, "elapsed_time": "1:30:55", "remaining_time": "7:52:49"}
5
+ {"current_steps": 60, "total_steps": 310, "loss": 1.2929, "reward": 13.2822, "lr": 9.784701678661045e-06, "epoch": 1.9, "percentage": 19.35, "elapsed_time": "1:49:13", "remaining_time": "7:35:05"}
6
+ {"current_steps": 70, "total_steps": 310, "loss": 1.0175, "reward": 15.9296, "lr": 9.598251102025463e-06, "epoch": 2.23, "percentage": 22.58, "elapsed_time": "2:02:46", "remaining_time": "7:00:57"}
7
+ {"current_steps": 80, "total_steps": 310, "loss": 1.5542, "reward": 14.864, "lr": 9.357139626751308e-06, "epoch": 2.55, "percentage": 25.81, "elapsed_time": "2:16:57", "remaining_time": "6:33:44"}
8
+ {"current_steps": 90, "total_steps": 310, "loss": 1.3869, "reward": 15.363, "lr": 9.064233422958078e-06, "epoch": 2.87, "percentage": 29.03, "elapsed_time": "2:28:56", "remaining_time": "6:04:04"}
9
+ {"current_steps": 100, "total_steps": 310, "loss": 1.0299, "reward": 16.2821, "lr": 8.723014361461633e-06, "epoch": 3.19, "percentage": 32.26, "elapsed_time": "2:40:28", "remaining_time": "5:37:00"}
10
+ {"current_steps": 110, "total_steps": 310, "loss": 0.8473, "reward": 15.4689, "lr": 8.337538623649237e-06, "epoch": 3.52, "percentage": 35.48, "elapsed_time": "2:51:52", "remaining_time": "5:12:29"}
11
+ {"current_steps": 120, "total_steps": 310, "loss": 0.717, "reward": 17.3127, "lr": 7.912388484339012e-06, "epoch": 3.84, "percentage": 38.71, "elapsed_time": "3:04:48", "remaining_time": "4:52:36"}
12
+ {"current_steps": 130, "total_steps": 310, "loss": 0.5949, "reward": 17.6693, "lr": 7.4526178407965396e-06, "epoch": 4.16, "percentage": 41.94, "elapsed_time": "3:17:31", "remaining_time": "4:33:30"}
13
+ {"current_steps": 140, "total_steps": 310, "loss": 0.6247, "reward": 16.8791, "lr": 6.963692135422872e-06, "epoch": 4.48, "percentage": 45.16, "elapsed_time": "3:30:46", "remaining_time": "4:15:56"}
14
+ {"current_steps": 150, "total_steps": 310, "loss": 0.7098, "reward": 17.0654, "lr": 6.451423386272312e-06, "epoch": 4.81, "percentage": 48.39, "elapsed_time": "3:44:04", "remaining_time": "3:59:00"}
15
+ {"current_steps": 160, "total_steps": 310, "loss": 0.4663, "reward": 17.231, "lr": 5.921901097713317e-06, "epoch": 5.13, "percentage": 51.61, "elapsed_time": "3:56:51", "remaining_time": "3:42:03"}
16
+ {"current_steps": 170, "total_steps": 310, "loss": 0.5734, "reward": 17.4632, "lr": 5.381419872519763e-06, "epoch": 5.45, "percentage": 54.84, "elapsed_time": "4:10:39", "remaining_time": "3:26:25"}
17
+ {"current_steps": 180, "total_steps": 310, "loss": 0.3885, "reward": 17.9993, "lr": 4.83640458589112e-06, "epoch": 5.77, "percentage": 58.06, "elapsed_time": "4:24:10", "remaining_time": "3:10:47"}
18
+ {"current_steps": 190, "total_steps": 310, "loss": 0.2806, "reward": 17.2133, "lr": 4.293334010882164e-06, "epoch": 6.1, "percentage": 61.29, "elapsed_time": "4:36:22", "remaining_time": "2:54:33"}
19
+ {"current_steps": 200, "total_steps": 310, "loss": 0.3708, "reward": 17.0068, "lr": 3.7586638031314182e-06, "epoch": 6.42, "percentage": 64.52, "elapsed_time": "4:49:14", "remaining_time": "2:39:04"}
20
+ {"current_steps": 210, "total_steps": 310, "loss": 0.4041, "reward": 17.7468, "lr": 3.2387497603938327e-06, "epoch": 6.74, "percentage": 67.74, "elapsed_time": "5:01:58", "remaining_time": "2:23:48"}
21
+ {"current_steps": 220, "total_steps": 310, "loss": 0.3516, "reward": 18.2141, "lr": 2.739772269116402e-06, "epoch": 7.06, "percentage": 70.97, "elapsed_time": "5:14:17", "remaining_time": "2:08:34"}
22
+ {"current_steps": 230, "total_steps": 310, "loss": 0.2936, "reward": 17.5244, "lr": 2.2676628361847834e-06, "epoch": 7.39, "percentage": 74.19, "elapsed_time": "5:26:45", "remaining_time": "1:53:39"}
23
+ {"current_steps": 240, "total_steps": 310, "loss": 0.2178, "reward": 17.7584, "lr": 1.8280335791817733e-06, "epoch": 7.71, "percentage": 77.42, "elapsed_time": "5:39:20", "remaining_time": "1:38:58"}
24
+ {"current_steps": 250, "total_steps": 310, "loss": 0.2559, "reward": 17.8964, "lr": 1.4261105133297693e-06, "epoch": 8.03, "percentage": 80.65, "elapsed_time": "5:51:04", "remaining_time": "1:24:15"}
25
+ {"current_steps": 260, "total_steps": 310, "loss": 0.2617, "reward": 18.0355, "lr": 1.0666714281569152e-06, "epoch": 8.35, "percentage": 83.87, "elapsed_time": "6:02:31", "remaining_time": "1:09:42"}
26
+ {"current_steps": 270, "total_steps": 310, "loss": 0.2195, "reward": 17.893, "lr": 7.539890923671061e-07, "epoch": 8.68, "percentage": 87.1, "elapsed_time": "6:14:20", "remaining_time": "0:55:27"}
27
+ {"current_steps": 280, "total_steps": 310, "loss": 0.1847, "reward": 17.9483, "lr": 4.917804620559202e-07, "epoch": 9.0, "percentage": 90.32, "elapsed_time": "6:25:56", "remaining_time": "0:41:21"}
28
+ {"current_steps": 290, "total_steps": 310, "loss": 0.1887, "reward": 18.637, "lr": 2.8316249605087386e-07, "epoch": 9.32, "percentage": 93.55, "elapsed_time": "6:38:05", "remaining_time": "0:27:27"}
29
+ {"current_steps": 300, "total_steps": 310, "loss": 0.2198, "reward": 17.2751, "lr": 1.3061510361333186e-07, "epoch": 9.65, "percentage": 96.77, "elapsed_time": "6:48:09", "remaining_time": "0:13:36"}
30
+ {"current_steps": 310, "total_steps": 310, "loss": 0.2581, "reward": 18.2905, "lr": 3.59516649547248e-08, "epoch": 9.97, "percentage": 100.0, "elapsed_time": "6:59:41", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": null,
5
+ "eval_steps": 500,
6
+ "global_step": 310,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.29,
13
+ "learning_rate": 2.8125e-06,
14
+ "loss": 3.9546,
15
+ "reward": 9.563,
16
+ "step": 9
17
+ },
18
+ {
19
+ "epoch": 0.61,
20
+ "learning_rate": 5.9375e-06,
21
+ "loss": 4.6766,
22
+ "reward": 8.9945,
23
+ "step": 19
24
+ },
25
+ {
26
+ "epoch": 0.94,
27
+ "learning_rate": 9.0625e-06,
28
+ "loss": 4.0729,
29
+ "reward": 9.3216,
30
+ "step": 29
31
+ },
32
+ {
33
+ "epoch": 1.26,
34
+ "learning_rate": 9.985430661522333e-06,
35
+ "loss": 3.0501,
36
+ "reward": 9.949,
37
+ "step": 39
38
+ },
39
+ {
40
+ "epoch": 1.58,
41
+ "learning_rate": 9.914274958326507e-06,
42
+ "loss": 1.148,
43
+ "reward": 11.1044,
44
+ "step": 49
45
+ },
46
+ {
47
+ "epoch": 1.9,
48
+ "learning_rate": 9.784701678661045e-06,
49
+ "loss": 1.2929,
50
+ "reward": 13.2822,
51
+ "step": 59
52
+ },
53
+ {
54
+ "epoch": 2.23,
55
+ "learning_rate": 9.598251102025463e-06,
56
+ "loss": 1.0175,
57
+ "reward": 15.9296,
58
+ "step": 69
59
+ },
60
+ {
61
+ "epoch": 2.55,
62
+ "learning_rate": 9.357139626751308e-06,
63
+ "loss": 1.5542,
64
+ "reward": 14.864,
65
+ "step": 79
66
+ },
67
+ {
68
+ "epoch": 2.87,
69
+ "learning_rate": 9.064233422958078e-06,
70
+ "loss": 1.3869,
71
+ "reward": 15.363,
72
+ "step": 89
73
+ },
74
+ {
75
+ "epoch": 3.19,
76
+ "learning_rate": 8.723014361461633e-06,
77
+ "loss": 1.0299,
78
+ "reward": 16.2821,
79
+ "step": 99
80
+ },
81
+ {
82
+ "epoch": 3.52,
83
+ "learning_rate": 8.337538623649237e-06,
84
+ "loss": 0.8473,
85
+ "reward": 15.4689,
86
+ "step": 109
87
+ },
88
+ {
89
+ "epoch": 3.84,
90
+ "learning_rate": 7.912388484339012e-06,
91
+ "loss": 0.717,
92
+ "reward": 17.3127,
93
+ "step": 119
94
+ },
95
+ {
96
+ "epoch": 4.16,
97
+ "learning_rate": 7.4526178407965396e-06,
98
+ "loss": 0.5949,
99
+ "reward": 17.6693,
100
+ "step": 129
101
+ },
102
+ {
103
+ "epoch": 4.48,
104
+ "learning_rate": 6.963692135422872e-06,
105
+ "loss": 0.6247,
106
+ "reward": 16.8791,
107
+ "step": 139
108
+ },
109
+ {
110
+ "epoch": 4.81,
111
+ "learning_rate": 6.451423386272312e-06,
112
+ "loss": 0.7098,
113
+ "reward": 17.0654,
114
+ "step": 149
115
+ },
116
+ {
117
+ "epoch": 5.13,
118
+ "learning_rate": 5.921901097713317e-06,
119
+ "loss": 0.4663,
120
+ "reward": 17.231,
121
+ "step": 159
122
+ },
123
+ {
124
+ "epoch": 5.45,
125
+ "learning_rate": 5.381419872519763e-06,
126
+ "loss": 0.5734,
127
+ "reward": 17.4632,
128
+ "step": 169
129
+ },
130
+ {
131
+ "epoch": 5.77,
132
+ "learning_rate": 4.83640458589112e-06,
133
+ "loss": 0.3885,
134
+ "reward": 17.9993,
135
+ "step": 179
136
+ },
137
+ {
138
+ "epoch": 6.1,
139
+ "learning_rate": 4.293334010882164e-06,
140
+ "loss": 0.2806,
141
+ "reward": 17.2133,
142
+ "step": 189
143
+ },
144
+ {
145
+ "epoch": 6.42,
146
+ "learning_rate": 3.7586638031314182e-06,
147
+ "loss": 0.3708,
148
+ "reward": 17.0068,
149
+ "step": 199
150
+ },
151
+ {
152
+ "epoch": 6.74,
153
+ "learning_rate": 3.2387497603938327e-06,
154
+ "loss": 0.4041,
155
+ "reward": 17.7468,
156
+ "step": 209
157
+ },
158
+ {
159
+ "epoch": 7.06,
160
+ "learning_rate": 2.739772269116402e-06,
161
+ "loss": 0.3516,
162
+ "reward": 18.2141,
163
+ "step": 219
164
+ },
165
+ {
166
+ "epoch": 7.39,
167
+ "learning_rate": 2.2676628361847834e-06,
168
+ "loss": 0.2936,
169
+ "reward": 17.5244,
170
+ "step": 229
171
+ },
172
+ {
173
+ "epoch": 7.71,
174
+ "learning_rate": 1.8280335791817733e-06,
175
+ "loss": 0.2178,
176
+ "reward": 17.7584,
177
+ "step": 239
178
+ },
179
+ {
180
+ "epoch": 8.03,
181
+ "learning_rate": 1.4261105133297693e-06,
182
+ "loss": 0.2559,
183
+ "reward": 17.8964,
184
+ "step": 249
185
+ },
186
+ {
187
+ "epoch": 8.35,
188
+ "learning_rate": 1.0666714281569152e-06,
189
+ "loss": 0.2617,
190
+ "reward": 18.0355,
191
+ "step": 259
192
+ },
193
+ {
194
+ "epoch": 8.68,
195
+ "learning_rate": 7.539890923671061e-07,
196
+ "loss": 0.2195,
197
+ "reward": 17.893,
198
+ "step": 269
199
+ },
200
+ {
201
+ "epoch": 9.0,
202
+ "learning_rate": 4.917804620559202e-07,
203
+ "loss": 0.1847,
204
+ "reward": 17.9483,
205
+ "step": 279
206
+ },
207
+ {
208
+ "epoch": 9.32,
209
+ "learning_rate": 2.8316249605087386e-07,
210
+ "loss": 0.1887,
211
+ "reward": 18.637,
212
+ "step": 289
213
+ },
214
+ {
215
+ "epoch": 9.65,
216
+ "learning_rate": 1.3061510361333186e-07,
217
+ "loss": 0.2198,
218
+ "reward": 17.2751,
219
+ "step": 299
220
+ },
221
+ {
222
+ "epoch": 9.97,
223
+ "learning_rate": 3.59516649547248e-08,
224
+ "loss": 0.2581,
225
+ "reward": 18.2905,
226
+ "step": 309
227
+ }
228
+ ],
229
+ "logging_steps": 500,
230
+ "max_steps": 310,
231
+ "num_input_tokens_seen": 0,
232
+ "num_train_epochs": 10.0,
233
+ "save_steps": 500,
234
+ "stateful_callbacks": {},
235
+ "total_flos": 0,
236
+ "train_batch_size": null,
237
+ "trial_name": null,
238
+ "trial_params": null
239
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16a23ae21c38266e71e578d342bfc4139cc1cc2a4b433d5167693ea8652df184
3
+ size 5624
training_loss.png ADDED
training_reward.png ADDED
value_head.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2761f60ca79ab817f864324eda7859de479dd405b18bd4feade95cb615a7c6f
3
+ size 16588