cryptalk3 commited on
Commit
21ac029
·
verified ·
1 Parent(s): c5de232

End of training

Browse files
README.md CHANGED
@@ -1,5 +1,5 @@
1
  ---
2
- base_model: Gensyn/Qwen2.5-0.5B-Instruct
3
  library_name: transformers
4
  model_name: gensyn-checkpoints-powerful_pouncing_buffalo
5
  tags:
@@ -8,13 +8,14 @@ tags:
8
  - grpo
9
  - gensyn
10
  - I am powerful pouncing buffalo
 
11
  - trl
12
  licence: license
13
  ---
14
 
15
  # Model Card for gensyn-checkpoints-powerful_pouncing_buffalo
16
 
17
- This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
18
  It has been trained using [TRL](https://github.com/huggingface/trl).
19
 
20
  ## Quick start
@@ -39,7 +40,7 @@ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing
39
 
40
  - TRL: 0.15.2
41
  - Transformers: 4.51.3
42
- - Pytorch: 2.5.1
43
  - Datasets: 3.5.0
44
  - Tokenizers: 0.21.1
45
 
 
1
  ---
2
+ base_model: Gensyn/Qwen2.5-1.5B-Instruct
3
  library_name: transformers
4
  model_name: gensyn-checkpoints-powerful_pouncing_buffalo
5
  tags:
 
8
  - grpo
9
  - gensyn
10
  - I am powerful pouncing buffalo
11
+ - unsloth
12
  - trl
13
  licence: license
14
  ---
15
 
16
  # Model Card for gensyn-checkpoints-powerful_pouncing_buffalo
17
 
18
+ This model is a fine-tuned version of [Gensyn/Qwen2.5-1.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-1.5B-Instruct).
19
  It has been trained using [TRL](https://github.com/huggingface/trl).
20
 
21
  ## Quick start
 
40
 
41
  - TRL: 0.15.2
42
  - Transformers: 4.51.3
43
+ - Pytorch: 2.6.0
44
  - Datasets: 3.5.0
45
  - Tokenizers: 0.21.1
46
 
adapter_config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "Gensyn/Qwen2.5-1.5B-Instruct",
5
+ "bias": "none",
6
+ "corda_config": null,
7
+ "eva_config": null,
8
+ "exclude_modules": null,
9
+ "fan_in_fan_out": false,
10
+ "inference_mode": true,
11
+ "init_lora_weights": true,
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 16,
17
+ "lora_bias": false,
18
+ "lora_dropout": 0,
19
+ "megatron_config": null,
20
+ "megatron_core": "megatron.core",
21
+ "modules_to_save": null,
22
+ "peft_type": "LORA",
23
+ "r": 16,
24
+ "rank_pattern": {},
25
+ "revision": null,
26
+ "target_modules": [
27
+ "up_proj",
28
+ "k_proj",
29
+ "o_proj",
30
+ "gate_proj",
31
+ "v_proj",
32
+ "q_proj",
33
+ "down_proj"
34
+ ],
35
+ "task_type": "CAUSAL_LM",
36
+ "trainable_token_indices": null,
37
+ "use_dora": false,
38
+ "use_rslora": false
39
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8480b14e41e8448b9d84c38d56f02d18204f12ff70daad578842a24ade01e114
3
+ size 73911112
all_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "total_flos": 0.0,
3
- "train_loss": 1.648552715778351e-05,
4
- "train_runtime": 203.3529,
5
- "train_samples": 28,
6
- "train_samples_per_second": 1.574,
7
- "train_steps_per_second": 0.098
8
  }
 
1
  {
2
  "total_flos": 0.0,
3
+ "train_loss": 7.103483881110151e-07,
4
+ "train_runtime": 1565.82,
5
+ "train_samples": 18,
6
+ "train_samples_per_second": 0.204,
7
+ "train_steps_per_second": 0.013
8
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "total_flos": 0.0,
3
- "train_loss": 1.648552715778351e-05,
4
- "train_runtime": 203.3529,
5
- "train_samples": 28,
6
- "train_samples_per_second": 1.574,
7
- "train_steps_per_second": 0.098
8
  }
 
1
  {
2
  "total_flos": 0.0,
3
+ "train_loss": 7.103483881110151e-07,
4
+ "train_runtime": 1565.82,
5
+ "train_samples": 18,
6
+ "train_samples_per_second": 0.204,
7
+ "train_steps_per_second": 0.013
8
  }
trainer_state.json CHANGED
@@ -2,7 +2,7 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 5.0,
6
  "eval_steps": 500,
7
  "global_step": 20,
8
  "is_hyper_param_search": false,
@@ -10,209 +10,209 @@
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
- "completion_length": 312.1875,
14
- "epoch": 0.5714285714285714,
15
- "grad_norm": 7.325751781463623,
16
- "kl": 0.0,
17
  "learning_rate": 5e-07,
18
  "loss": 0.0,
19
- "reward": 1.6316200233995914,
20
- "reward_std": 0.9343204437755048,
21
- "rewards/concensus_correctness_reward_func": 0.18800000101327896,
22
  "rewards/consensus_reward_func": 0.25,
23
  "rewards/cumulative_reward_2": 0.0,
24
- "rewards/final_correctness_reward_func": 0.0625,
25
- "rewards/question_recreation_reward_func": 0.4644637518795207,
26
  "rewards/soft_format_reward_func": 0.0,
27
- "rewards/strict_format_reward_func": 0.0,
28
- "rewards/xmlcount_reward_func": 0.6666562519967556,
29
  "step": 2
30
  },
31
  {
32
- "completion_length": 248.29166666666666,
33
- "epoch": 1.0,
34
- "grad_norm": 5.5042009353637695,
35
- "kl": 0.0016824240350009252,
36
  "learning_rate": 4.864543104251586e-07,
37
  "loss": 0.0,
38
- "reward": 1.9117889901002247,
39
- "reward_std": 0.8826162366506954,
40
- "rewards/concensus_correctness_reward_func": 0.16300000250339508,
41
- "rewards/consensus_reward_func": 0.16666666666666666,
42
  "rewards/cumulative_reward_2": 0.0,
43
- "rewards/final_correctness_reward_func": 0.08333333333333333,
44
- "rewards/question_recreation_reward_func": 0.5494973274568716,
45
  "rewards/soft_format_reward_func": 0.0,
46
  "rewards/strict_format_reward_func": 0.0,
47
- "rewards/xmlcount_reward_func": 0.9492916663487753,
48
  "step": 4
49
  },
50
  {
51
- "completion_length": 202.0625,
52
- "epoch": 1.5714285714285714,
53
- "grad_norm": 8.567811965942383,
54
- "kl": 0.00944961684399459,
55
  "learning_rate": 4.472851273490984e-07,
56
  "loss": 0.0,
57
- "reward": 1.7320603746920824,
58
- "reward_std": 1.027712611365132,
59
- "rewards/concensus_correctness_reward_func": 0.2104375008493662,
60
- "rewards/consensus_reward_func": 0.25,
61
  "rewards/cumulative_reward_2": 0.0,
62
- "rewards/final_correctness_reward_func": 0.1875,
63
- "rewards/question_recreation_reward_func": 0.4185916268033907,
64
  "rewards/soft_format_reward_func": 0.0,
65
  "rewards/strict_format_reward_func": 0.0,
66
- "rewards/xmlcount_reward_func": 0.6655312478542328,
67
  "step": 6
68
  },
69
  {
70
- "completion_length": 174.5,
71
- "epoch": 2.0,
72
- "grad_norm": 5.557609558105469,
73
- "kl": 0.005153165373485535,
74
  "learning_rate": 3.867370395306068e-07,
75
  "loss": 0.0,
76
- "reward": 2.4640907049179077,
77
- "reward_std": 1.0979739256824057,
78
- "rewards/concensus_correctness_reward_func": 0.41366667052110034,
79
- "rewards/consensus_reward_func": 0.5,
80
  "rewards/cumulative_reward_2": 0.0,
81
- "rewards/final_correctness_reward_func": 0.08333333333333333,
82
- "rewards/question_recreation_reward_func": 0.4825073642035325,
83
  "rewards/soft_format_reward_func": 0.0,
84
- "rewards/strict_format_reward_func": 0.0,
85
- "rewards/xmlcount_reward_func": 0.9845833430687586,
86
  "step": 8
87
  },
88
  {
89
- "completion_length": 291.875,
90
- "epoch": 2.571428571428571,
91
- "grad_norm": 8.104419708251953,
92
- "kl": 0.012241542688570917,
93
  "learning_rate": 3.1137137178519977e-07,
94
  "loss": 0.0,
95
- "reward": 2.0051864236593246,
96
- "reward_std": 1.0001586927101016,
97
- "rewards/concensus_correctness_reward_func": 0.3031250014901161,
98
- "rewards/consensus_reward_func": 0.25,
99
  "rewards/cumulative_reward_2": 0.0,
100
- "rewards/final_correctness_reward_func": 0.1875,
101
- "rewards/question_recreation_reward_func": 0.5442801603348926,
102
  "rewards/soft_format_reward_func": 0.0,
103
  "rewards/strict_format_reward_func": 0.0,
104
- "rewards/xmlcount_reward_func": 0.7202812451869249,
105
  "step": 10
106
  },
107
  {
108
- "completion_length": 210.0,
109
- "epoch": 3.0,
110
- "grad_norm": 6.57753849029541,
111
- "kl": 0.021596300435097266,
112
  "learning_rate": 2.2935516363191693e-07,
113
  "loss": 0.0,
114
- "reward": 2.404164135456085,
115
- "reward_std": 1.113425424322486,
116
- "rewards/concensus_correctness_reward_func": 0.4017500082651774,
117
- "rewards/consensus_reward_func": 0.4166666666666667,
118
  "rewards/cumulative_reward_2": 0.0,
119
  "rewards/final_correctness_reward_func": 0.0,
120
- "rewards/question_recreation_reward_func": 0.71028916289409,
121
  "rewards/soft_format_reward_func": 0.0,
122
- "rewards/strict_format_reward_func": 0.0,
123
- "rewards/xmlcount_reward_func": 0.875458337366581,
124
  "step": 12
125
  },
126
  {
127
- "completion_length": 239.1875,
128
- "epoch": 3.571428571428571,
129
- "grad_norm": 7.574004173278809,
130
- "kl": 0.010846115357708186,
131
  "learning_rate": 1.4957614383675767e-07,
132
  "loss": 0.0,
133
- "reward": 1.908155683428049,
134
- "reward_std": 1.323273050133139,
135
- "rewards/concensus_correctness_reward_func": 0.27850000001490116,
136
- "rewards/consensus_reward_func": 0.25,
137
  "rewards/cumulative_reward_2": 0.0,
138
- "rewards/final_correctness_reward_func": 0.0625,
139
- "rewards/question_recreation_reward_func": 0.4894369235262275,
140
- "rewards/soft_format_reward_func": 0.0,
141
  "rewards/strict_format_reward_func": 0.0,
142
- "rewards/xmlcount_reward_func": 0.8277187719941139,
143
  "step": 14
144
  },
145
  {
146
- "completion_length": 247.79166666666666,
147
- "epoch": 4.0,
148
- "grad_norm": 6.369115352630615,
149
- "kl": 0.10813957840825121,
150
  "learning_rate": 8.067960709356478e-08,
151
- "loss": 0.0001,
152
- "reward": 1.8096545437971752,
153
- "reward_std": 0.9561568250258764,
154
- "rewards/concensus_correctness_reward_func": 0.16191666821638742,
155
- "rewards/consensus_reward_func": 0.16666666666666666,
156
  "rewards/cumulative_reward_2": 0.0,
157
- "rewards/final_correctness_reward_func": 0.08333333333333333,
158
- "rewards/question_recreation_reward_func": 0.5189461906750997,
159
- "rewards/soft_format_reward_func": 0.0,
160
  "rewards/strict_format_reward_func": 0.0,
161
- "rewards/xmlcount_reward_func": 0.8787916700045267,
162
  "step": 16
163
  },
164
  {
165
- "completion_length": 216.6875,
166
- "epoch": 4.571428571428571,
167
- "grad_norm": 7.487825393676758,
168
- "kl": 0.01603295307722874,
169
  "learning_rate": 3.013156219837776e-08,
170
  "loss": 0.0,
171
- "reward": 2.2358960770070553,
172
- "reward_std": 1.0591734377667308,
173
- "rewards/concensus_correctness_reward_func": 0.3711875043809414,
174
- "rewards/consensus_reward_func": 0.4375,
175
  "rewards/cumulative_reward_2": 0.0,
176
- "rewards/final_correctness_reward_func": 0.0,
177
- "rewards/question_recreation_reward_func": 0.5273960698395967,
178
  "rewards/soft_format_reward_func": 0.0,
179
- "rewards/strict_format_reward_func": 0.0,
180
- "rewards/xmlcount_reward_func": 0.8998125158250332,
181
  "step": 18
182
  },
183
  {
184
- "completion_length": 225.54166666666666,
185
- "epoch": 5.0,
186
- "grad_norm": 4.96890115737915,
187
- "kl": 0.018138774127388995,
188
  "learning_rate": 3.4096741493194193e-09,
189
  "loss": 0.0,
190
- "reward": 1.6644823551177979,
191
- "reward_std": 1.286612228800853,
192
- "rewards/concensus_correctness_reward_func": 0.16183333098888397,
193
- "rewards/consensus_reward_func": 0.16666666666666666,
194
  "rewards/cumulative_reward_2": 0.0,
195
  "rewards/final_correctness_reward_func": 0.25,
196
- "rewards/question_recreation_reward_func": 0.5021490190799037,
197
  "rewards/soft_format_reward_func": 0.0,
198
  "rewards/strict_format_reward_func": 0.0,
199
- "rewards/xmlcount_reward_func": 0.5838333169619242,
200
  "step": 20
201
  },
202
  {
203
- "epoch": 5.0,
204
  "step": 20,
205
  "total_flos": 0.0,
206
- "train_loss": 1.648552715778351e-05,
207
- "train_runtime": 203.3529,
208
- "train_samples_per_second": 1.574,
209
- "train_steps_per_second": 0.098
210
  }
211
  ],
212
  "logging_steps": 2,
213
  "max_steps": 20,
214
  "num_input_tokens_seen": 0,
215
- "num_train_epochs": 7,
216
  "save_steps": 25,
217
  "stateful_callbacks": {
218
  "TrainerControl": {
@@ -227,7 +227,7 @@
227
  }
228
  },
229
  "total_flos": 0.0,
230
- "train_batch_size": 2,
231
  "trial_name": null,
232
  "trial_params": null
233
  }
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 4.0,
6
  "eval_steps": 500,
7
  "global_step": 20,
8
  "is_hyper_param_search": false,
 
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
+ "completion_length": 357.1875,
14
+ "epoch": 0.4444444444444444,
15
+ "grad_norm": 0.7581892013549805,
16
+ "kl": 0.0007120595255400985,
17
  "learning_rate": 5e-07,
18
  "loss": 0.0,
19
+ "reward": 1.4256322383880615,
20
+ "reward_std": 1.2912949696183205,
21
+ "rewards/concensus_correctness_reward_func": 0.07556249853223562,
22
  "rewards/consensus_reward_func": 0.25,
23
  "rewards/cumulative_reward_2": 0.0,
24
+ "rewards/final_correctness_reward_func": 0.25,
25
+ "rewards/question_recreation_reward_func": 0.7241634875535965,
26
  "rewards/soft_format_reward_func": 0.0,
27
+ "rewards/strict_format_reward_func": 0.015625,
28
+ "rewards/xmlcount_reward_func": 0.11028125509619713,
29
  "step": 2
30
  },
31
  {
32
+ "completion_length": 363.0625,
33
+ "epoch": 0.8888888888888888,
34
+ "grad_norm": 0.4985937774181366,
35
+ "kl": 0.0008577058833907358,
36
  "learning_rate": 4.864543104251586e-07,
37
  "loss": 0.0,
38
+ "reward": 1.1555784791707993,
39
+ "reward_std": 0.9097523726522923,
40
+ "rewards/concensus_correctness_reward_func": 0.18950000405311584,
41
+ "rewards/consensus_reward_func": 0.25,
42
  "rewards/cumulative_reward_2": 0.0,
43
+ "rewards/final_correctness_reward_func": 0.0,
44
+ "rewards/question_recreation_reward_func": 0.35520345717668533,
45
  "rewards/soft_format_reward_func": 0.0,
46
  "rewards/strict_format_reward_func": 0.0,
47
+ "rewards/xmlcount_reward_func": 0.36087501142174006,
48
  "step": 4
49
  },
50
  {
51
+ "completion_length": 349.2916666666667,
52
+ "epoch": 1.2222222222222223,
53
+ "grad_norm": 0.514335036277771,
54
+ "kl": 0.0007775216557395955,
55
  "learning_rate": 4.472851273490984e-07,
56
  "loss": 0.0,
57
+ "reward": 1.3893702427546184,
58
+ "reward_std": 1.0876383855938911,
59
+ "rewards/concensus_correctness_reward_func": 0.0,
60
+ "rewards/consensus_reward_func": 0.5,
61
  "rewards/cumulative_reward_2": 0.0,
62
+ "rewards/final_correctness_reward_func": 0.0,
63
+ "rewards/question_recreation_reward_func": 0.653828593591849,
64
  "rewards/soft_format_reward_func": 0.0,
65
  "rewards/strict_format_reward_func": 0.0,
66
+ "rewards/xmlcount_reward_func": 0.23554165661334991,
67
  "step": 6
68
  },
69
  {
70
+ "completion_length": 346.09375,
71
+ "epoch": 1.6666666666666665,
72
+ "grad_norm": 0.6408029794692993,
73
+ "kl": 0.0008339681153302081,
74
  "learning_rate": 3.867370395306068e-07,
75
  "loss": 0.0,
76
+ "reward": 1.8015710562467575,
77
+ "reward_std": 1.0655995905399323,
78
+ "rewards/concensus_correctness_reward_func": 0.19812499731779099,
79
+ "rewards/consensus_reward_func": 0.375,
80
  "rewards/cumulative_reward_2": 0.0,
81
+ "rewards/final_correctness_reward_func": 0.0,
82
+ "rewards/question_recreation_reward_func": 0.5710397940129042,
83
  "rewards/soft_format_reward_func": 0.0,
84
+ "rewards/strict_format_reward_func": 0.046875,
85
+ "rewards/xmlcount_reward_func": 0.6105312574654818,
86
  "step": 8
87
  },
88
  {
89
+ "completion_length": 418.0,
90
+ "epoch": 2.0,
91
+ "grad_norm": 0.3388269543647766,
92
+ "kl": 0.0008536583142510304,
93
  "learning_rate": 3.1137137178519977e-07,
94
  "loss": 0.0,
95
+ "reward": 1.3034771978855133,
96
+ "reward_std": 0.9530518700679144,
97
+ "rewards/concensus_correctness_reward_func": 0.0,
98
+ "rewards/consensus_reward_func": 0.3333333333333333,
99
  "rewards/cumulative_reward_2": 0.0,
100
+ "rewards/final_correctness_reward_func": 0.08333333333333333,
101
+ "rewards/question_recreation_reward_func": 0.7229771912097931,
102
  "rewards/soft_format_reward_func": 0.0,
103
  "rewards/strict_format_reward_func": 0.0,
104
+ "rewards/xmlcount_reward_func": 0.16383333007494608,
105
  "step": 10
106
  },
107
  {
108
+ "completion_length": 370.71875,
109
+ "epoch": 2.4444444444444446,
110
+ "grad_norm": 0.5266546010971069,
111
+ "kl": 0.000776094962930074,
112
  "learning_rate": 2.2935516363191693e-07,
113
  "loss": 0.0,
114
+ "reward": 1.1951591446995735,
115
+ "reward_std": 0.9491450227797031,
116
+ "rewards/concensus_correctness_reward_func": 0.023125000298023224,
117
+ "rewards/consensus_reward_func": 0.375,
118
  "rewards/cumulative_reward_2": 0.0,
119
  "rewards/final_correctness_reward_func": 0.0,
120
+ "rewards/question_recreation_reward_func": 0.4554091515019536,
121
  "rewards/soft_format_reward_func": 0.0,
122
+ "rewards/strict_format_reward_func": 0.015625,
123
+ "rewards/xmlcount_reward_func": 0.3260000143200159,
124
  "step": 12
125
  },
126
  {
127
+ "completion_length": 389.0,
128
+ "epoch": 2.888888888888889,
129
+ "grad_norm": 0.4685328006744385,
130
+ "kl": 0.0008191447559511289,
131
  "learning_rate": 1.4957614383675767e-07,
132
  "loss": 0.0,
133
+ "reward": 1.0190585181117058,
134
+ "reward_std": 0.9665377400815487,
135
+ "rewards/concensus_correctness_reward_func": 0.01556249987334013,
136
+ "rewards/consensus_reward_func": 0.125,
137
  "rewards/cumulative_reward_2": 0.0,
138
+ "rewards/final_correctness_reward_func": 0.0,
139
+ "rewards/question_recreation_reward_func": 0.6400585174560547,
140
+ "rewards/soft_format_reward_func": 0.015625,
141
  "rewards/strict_format_reward_func": 0.0,
142
+ "rewards/xmlcount_reward_func": 0.2228124998509884,
143
  "step": 14
144
  },
145
  {
146
+ "completion_length": 377.4166666666667,
147
+ "epoch": 3.2222222222222223,
148
+ "grad_norm": 0.42430800199508667,
149
+ "kl": 0.0007815616457567861,
150
  "learning_rate": 8.067960709356478e-08,
151
+ "loss": 0.0,
152
+ "reward": 1.6954653511444728,
153
+ "reward_std": 1.16114375491937,
154
+ "rewards/concensus_correctness_reward_func": 0.16033333539962769,
155
+ "rewards/consensus_reward_func": 0.5,
156
  "rewards/cumulative_reward_2": 0.0,
157
+ "rewards/final_correctness_reward_func": 0.4166666666666667,
158
+ "rewards/question_recreation_reward_func": 0.47421534235278767,
159
+ "rewards/soft_format_reward_func": 0.020833333333333332,
160
  "rewards/strict_format_reward_func": 0.0,
161
+ "rewards/xmlcount_reward_func": 0.12341668456792831,
162
  "step": 16
163
  },
164
  {
165
+ "completion_length": 335.34375,
166
+ "epoch": 3.6666666666666665,
167
+ "grad_norm": 0.8147290945053101,
168
+ "kl": 0.0007904403100837953,
169
  "learning_rate": 3.013156219837776e-08,
170
  "loss": 0.0,
171
+ "reward": 1.8439632058143616,
172
+ "reward_std": 1.37743728607893,
173
+ "rewards/concensus_correctness_reward_func": 0.07556249853223562,
174
+ "rewards/consensus_reward_func": 0.5625,
175
  "rewards/cumulative_reward_2": 0.0,
176
+ "rewards/final_correctness_reward_func": 0.125,
177
+ "rewards/question_recreation_reward_func": 0.6351507157087326,
178
  "rewards/soft_format_reward_func": 0.0,
179
+ "rewards/strict_format_reward_func": 0.03125,
180
+ "rewards/xmlcount_reward_func": 0.41450000926852226,
181
  "step": 18
182
  },
183
  {
184
+ "completion_length": 468.5416666666667,
185
+ "epoch": 4.0,
186
+ "grad_norm": 0.2882421016693115,
187
+ "kl": 0.0007724291062913835,
188
  "learning_rate": 3.4096741493194193e-09,
189
  "loss": 0.0,
190
+ "reward": 1.4719520608584087,
191
+ "reward_std": 1.2038606504599254,
192
+ "rewards/concensus_correctness_reward_func": 0.046666666865348816,
193
+ "rewards/consensus_reward_func": 0.3333333333333333,
194
  "rewards/cumulative_reward_2": 0.0,
195
  "rewards/final_correctness_reward_func": 0.25,
196
+ "rewards/question_recreation_reward_func": 0.5704936931530634,
197
  "rewards/soft_format_reward_func": 0.0,
198
  "rewards/strict_format_reward_func": 0.0,
199
+ "rewards/xmlcount_reward_func": 0.27145833025376004,
200
  "step": 20
201
  },
202
  {
203
+ "epoch": 4.0,
204
  "step": 20,
205
  "total_flos": 0.0,
206
+ "train_loss": 7.103483881110151e-07,
207
+ "train_runtime": 1565.82,
208
+ "train_samples_per_second": 0.204,
209
+ "train_steps_per_second": 0.013
210
  }
211
  ],
212
  "logging_steps": 2,
213
  "max_steps": 20,
214
  "num_input_tokens_seen": 0,
215
+ "num_train_epochs": 5,
216
  "save_steps": 25,
217
  "stateful_callbacks": {
218
  "TrainerControl": {
 
227
  }
228
  },
229
  "total_flos": 0.0,
230
+ "train_batch_size": 4,
231
  "trial_name": null,
232
  "trial_params": null
233
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0b9cd0859109d11e049f1b3691f70670cc9172725010e006f36ea5f81f0e6bb3
3
- size 5880
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbdc209ff9c1efe1e8f89f709417e3b6eed5af4a40d1a90df8714ca1663aeb93
3
+ size 5944