ztysefd commited on
Commit
d242491
·
verified ·
1 Parent(s): 669be80

Model save

Browse files
README.md CHANGED
@@ -1,10 +1,8 @@
1
  ---
2
- datasets: leonardPKU/GEOQA_R1V_Train_8K
3
  library_name: transformers
4
  model_name: Qwen2-VL-2B-Instruct-SFT
5
  tags:
6
  - generated_from_trainer
7
- - R1-V
8
  - trl
9
  - sft
10
  licence: license
@@ -12,7 +10,7 @@ licence: license
12
 
13
  # Model Card for Qwen2-VL-2B-Instruct-SFT
14
 
15
- This model is a fine-tuned version of [None](https://huggingface.co/None) on the [leonardPKU/GEOQA_R1V_Train_8K](https://huggingface.co/datasets/leonardPKU/GEOQA_R1V_Train_8K) dataset.
16
  It has been trained using [TRL](https://github.com/huggingface/trl).
17
 
18
  ## Quick start
@@ -28,7 +26,7 @@ print(output["generated_text"])
28
 
29
  ## Training procedure
30
 
31
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/502023330085-nanjing-university/huggingface/runs/vy5m53zx)
32
 
33
 
34
  This model was trained with SFT.
 
1
  ---
 
2
  library_name: transformers
3
  model_name: Qwen2-VL-2B-Instruct-SFT
4
  tags:
5
  - generated_from_trainer
 
6
  - trl
7
  - sft
8
  licence: license
 
10
 
11
  # Model Card for Qwen2-VL-2B-Instruct-SFT
12
 
13
+ This model is a fine-tuned version of [None](https://huggingface.co/None).
14
  It has been trained using [TRL](https://github.com/huggingface/trl).
15
 
16
  ## Quick start
 
26
 
27
  ## Training procedure
28
 
29
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/502023330085-nanjing-university/huggingface/runs/iug0psyq)
30
 
31
 
32
  This model was trained with SFT.
all_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 0.9880478087649402,
3
- "total_flos": 1.1032769987608576e+17,
4
- "train_loss": 0.4890208907665745,
5
- "train_runtime": 400.2546,
6
- "train_samples": 8031,
7
- "train_samples_per_second": 20.065,
8
- "train_steps_per_second": 0.155
9
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "total_flos": 1.0069004076135219e+18,
4
+ "train_loss": 0.04387825021601769,
5
+ "train_runtime": 3854.1911,
6
+ "train_samples": 70000,
7
+ "train_samples_per_second": 18.162,
8
+ "train_steps_per_second": 0.142
9
  }
config.json CHANGED
@@ -32,7 +32,7 @@
32
  "tie_word_embeddings": false,
33
  "torch_dtype": "bfloat16",
34
  "transformers_version": "4.49.0.dev0",
35
- "use_cache": true,
36
  "use_sliding_window": false,
37
  "video_token_id": 151656,
38
  "vision_config": {
 
32
  "tie_word_embeddings": false,
33
  "torch_dtype": "bfloat16",
34
  "transformers_version": "4.49.0.dev0",
35
+ "use_cache": false,
36
  "use_sliding_window": false,
37
  "video_token_id": 151656,
38
  "vision_config": {
model-00001-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4f5f28f933dd1da05a4eba62778f6c8f25b8c02c6819b35ac08b4d3f7523856
3
  size 4966659944
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fcbac1ad88c04a196f48d0bae145251d0af7a8de8634b53d9bdac8933e1e6a51
3
  size 4966659944
model-00002-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a67281fe16e0273a12f7c1dac91963027b3095b6279542029249d42f45c93ad7
3
  size 4991495816
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8096bce14807c410a5f75d39e1fa4ff324294d9fbc07ca117193a9263c10d7b5
3
  size 4991495816
model-00003-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3adf14f19964dc22a7fe49a9710b1b9ec93e5bb7c04e191a0442d39cfd532e2c
3
  size 4932751040
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c5fd037538512b298b9e4975fe42be4afb9f746d5085c9d525313ed0e441bf8
3
  size 4932751040
model-00004-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d0e57311ff6f36fa72a52df267bad8c8825e6471ca3ddf36d83de8d38e1ee52c
3
  size 1691924384
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fca744913c69ab1588604db7aa4b6d3d966db314fc42861522e0b00386ef679
3
  size 1691924384
tokenizer_config.json CHANGED
@@ -138,7 +138,6 @@
138
  "model_max_length": 32768,
139
  "pad_token": "<|endoftext|>",
140
  "padding_side": "left",
141
- "processor_class": "Qwen2VLProcessor",
142
  "split_special_tokens": false,
143
  "tokenizer_class": "Qwen2Tokenizer",
144
  "unk_token": null
 
138
  "model_max_length": 32768,
139
  "pad_token": "<|endoftext|>",
140
  "padding_side": "left",
 
141
  "split_special_tokens": false,
142
  "tokenizer_class": "Qwen2Tokenizer",
143
  "unk_token": null
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 0.9880478087649402,
3
- "total_flos": 1.1032769987608576e+17,
4
- "train_loss": 0.4890208907665745,
5
- "train_runtime": 400.2546,
6
- "train_samples": 8031,
7
- "train_samples_per_second": 20.065,
8
- "train_steps_per_second": 0.155
9
  }
 
1
  {
2
+ "epoch": 1.0,
3
+ "total_flos": 1.0069004076135219e+18,
4
+ "train_loss": 0.04387825021601769,
5
+ "train_runtime": 3854.1911,
6
+ "train_samples": 70000,
7
+ "train_samples_per_second": 18.162,
8
+ "train_steps_per_second": 0.142
9
  }
trainer_state.json CHANGED
@@ -1,109 +1,788 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 0.9880478087649402,
5
  "eval_steps": 500,
6
- "global_step": 62,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.0796812749003984,
13
- "grad_norm": 32.48837661743164,
14
- "learning_rate": 1.4285714285714287e-05,
15
- "loss": 1.9389,
16
  "step": 5
17
  },
18
  {
19
- "epoch": 0.1593625498007968,
20
- "grad_norm": 15.308554649353027,
21
- "learning_rate": 1.9853538358476933e-05,
22
- "loss": 0.5057,
23
  "step": 10
24
  },
25
  {
26
- "epoch": 0.23904382470119523,
27
- "grad_norm": 2.56449818611145,
28
- "learning_rate": 1.8973984286913584e-05,
29
- "loss": 0.5649,
30
  "step": 15
31
  },
32
  {
33
- "epoch": 0.3187250996015936,
34
- "grad_norm": 2.329458713531494,
35
- "learning_rate": 1.736741137876405e-05,
36
- "loss": 0.3706,
37
  "step": 20
38
  },
39
  {
40
- "epoch": 0.398406374501992,
41
- "grad_norm": 1.0648982524871826,
42
- "learning_rate": 1.5163974616389621e-05,
43
- "loss": 0.3278,
44
  "step": 25
45
  },
46
  {
47
- "epoch": 0.47808764940239046,
48
- "grad_norm": 1.1163208484649658,
49
- "learning_rate": 1.2542183341934873e-05,
50
- "loss": 0.3351,
51
  "step": 30
52
  },
53
  {
54
- "epoch": 0.5577689243027888,
55
- "grad_norm": 1.006535530090332,
56
- "learning_rate": 9.71443949206304e-06,
57
- "loss": 0.3183,
58
  "step": 35
59
  },
60
  {
61
- "epoch": 0.6374501992031872,
62
- "grad_norm": 1.0628993511199951,
63
- "learning_rate": 6.909830056250527e-06,
64
- "loss": 0.3149,
65
  "step": 40
66
  },
67
  {
68
- "epoch": 0.7171314741035857,
69
- "grad_norm": 0.9992952346801758,
70
- "learning_rate": 4.355567811332311e-06,
71
- "loss": 0.3166,
72
  "step": 45
73
  },
74
  {
75
- "epoch": 0.796812749003984,
76
- "grad_norm": 1.3005669116973877,
77
- "learning_rate": 2.2585838936091753e-06,
78
- "loss": 0.3017,
79
  "step": 50
80
  },
81
  {
82
- "epoch": 0.8764940239043825,
83
- "grad_norm": 0.8207844495773315,
84
- "learning_rate": 7.887634688515e-07,
85
- "loss": 0.3037,
86
  "step": 55
87
  },
88
  {
89
- "epoch": 0.9561752988047809,
90
- "grad_norm": 0.8011977076530457,
91
- "learning_rate": 6.51826465144978e-08,
92
- "loss": 0.311,
93
  "step": 60
94
  },
95
  {
96
- "epoch": 0.9880478087649402,
97
- "step": 62,
98
- "total_flos": 1.1032769987608576e+17,
99
- "train_loss": 0.4890208907665745,
100
- "train_runtime": 400.2546,
101
- "train_samples_per_second": 20.065,
102
- "train_steps_per_second": 0.155
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
103
  }
104
  ],
105
  "logging_steps": 5,
106
- "max_steps": 62,
107
  "num_input_tokens_seen": 0,
108
  "num_train_epochs": 1,
109
  "save_steps": 500,
@@ -119,7 +798,7 @@
119
  "attributes": {}
120
  }
121
  },
122
- "total_flos": 1.1032769987608576e+17,
123
  "train_batch_size": 4,
124
  "trial_name": null,
125
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
  "eval_steps": 500,
6
+ "global_step": 547,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.009140767824497258,
13
+ "grad_norm": 52.43793869018555,
14
+ "learning_rate": 1.8181818181818183e-06,
15
+ "loss": 2.7415,
16
  "step": 5
17
  },
18
  {
19
+ "epoch": 0.018281535648994516,
20
+ "grad_norm": 25.07411003112793,
21
+ "learning_rate": 3.6363636363636366e-06,
22
+ "loss": 1.4622,
23
  "step": 10
24
  },
25
  {
26
+ "epoch": 0.027422303473491772,
27
+ "grad_norm": 4.931793212890625,
28
+ "learning_rate": 5.4545454545454545e-06,
29
+ "loss": 0.1586,
30
  "step": 15
31
  },
32
  {
33
+ "epoch": 0.03656307129798903,
34
+ "grad_norm": 0.516838550567627,
35
+ "learning_rate": 7.272727272727273e-06,
36
+ "loss": 0.0265,
37
  "step": 20
38
  },
39
  {
40
+ "epoch": 0.04570383912248629,
41
+ "grad_norm": 14.795572280883789,
42
+ "learning_rate": 9.090909090909091e-06,
43
+ "loss": 0.068,
44
  "step": 25
45
  },
46
  {
47
+ "epoch": 0.054844606946983544,
48
+ "grad_norm": 9.015242576599121,
49
+ "learning_rate": 1.0909090909090909e-05,
50
+ "loss": 0.2448,
51
  "step": 30
52
  },
53
  {
54
+ "epoch": 0.06398537477148081,
55
+ "grad_norm": 0.29367485642433167,
56
+ "learning_rate": 1.2727272727272728e-05,
57
+ "loss": 0.0298,
58
  "step": 35
59
  },
60
  {
61
+ "epoch": 0.07312614259597806,
62
+ "grad_norm": 0.6809912919998169,
63
+ "learning_rate": 1.4545454545454546e-05,
64
+ "loss": 0.0029,
65
  "step": 40
66
  },
67
  {
68
+ "epoch": 0.08226691042047532,
69
+ "grad_norm": 0.5797951817512512,
70
+ "learning_rate": 1.6363636363636366e-05,
71
+ "loss": 0.0031,
72
  "step": 45
73
  },
74
  {
75
+ "epoch": 0.09140767824497258,
76
+ "grad_norm": 0.11719339340925217,
77
+ "learning_rate": 1.8181818181818182e-05,
78
+ "loss": 0.0025,
79
  "step": 50
80
  },
81
  {
82
+ "epoch": 0.10054844606946983,
83
+ "grad_norm": 0.3921687602996826,
84
+ "learning_rate": 2e-05,
85
+ "loss": 0.002,
86
  "step": 55
87
  },
88
  {
89
+ "epoch": 0.10968921389396709,
90
+ "grad_norm": 1.4710060358047485,
91
+ "learning_rate": 1.9994903844605973e-05,
92
+ "loss": 0.0019,
93
  "step": 60
94
  },
95
  {
96
+ "epoch": 0.11882998171846434,
97
+ "grad_norm": 1.201702356338501,
98
+ "learning_rate": 1.9979620572583846e-05,
99
+ "loss": 0.0013,
100
+ "step": 65
101
+ },
102
+ {
103
+ "epoch": 0.12797074954296161,
104
+ "grad_norm": 0.15323181450366974,
105
+ "learning_rate": 1.995416576111945e-05,
106
+ "loss": 0.0031,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 0.13711151736745886,
111
+ "grad_norm": 0.46892285346984863,
112
+ "learning_rate": 1.9918565354547738e-05,
113
+ "loss": 0.0028,
114
+ "step": 75
115
+ },
116
+ {
117
+ "epoch": 0.14625228519195613,
118
+ "grad_norm": 0.2431010752916336,
119
+ "learning_rate": 1.9872855637909506e-05,
120
+ "loss": 0.0017,
121
+ "step": 80
122
+ },
123
+ {
124
+ "epoch": 0.15539305301645337,
125
+ "grad_norm": 0.2643240690231323,
126
+ "learning_rate": 1.9817083199968552e-05,
127
+ "loss": 0.0012,
128
+ "step": 85
129
+ },
130
+ {
131
+ "epoch": 0.16453382084095064,
132
+ "grad_norm": 0.22663037478923798,
133
+ "learning_rate": 1.9751304885726967e-05,
134
+ "loss": 0.0007,
135
+ "step": 90
136
+ },
137
+ {
138
+ "epoch": 0.1736745886654479,
139
+ "grad_norm": 0.14844365417957306,
140
+ "learning_rate": 1.9675587738486935e-05,
141
+ "loss": 0.0003,
142
+ "step": 95
143
+ },
144
+ {
145
+ "epoch": 0.18281535648994515,
146
+ "grad_norm": 0.4369293749332428,
147
+ "learning_rate": 1.9590008931518133e-05,
148
+ "loss": 0.0005,
149
+ "step": 100
150
+ },
151
+ {
152
+ "epoch": 0.19195612431444242,
153
+ "grad_norm": 0.8838499784469604,
154
+ "learning_rate": 1.9494655689400294e-05,
155
+ "loss": 0.0035,
156
+ "step": 105
157
+ },
158
+ {
159
+ "epoch": 0.20109689213893966,
160
+ "grad_norm": 1.3618451356887817,
161
+ "learning_rate": 1.9389625199121264e-05,
162
+ "loss": 0.0049,
163
+ "step": 110
164
+ },
165
+ {
166
+ "epoch": 0.21023765996343693,
167
+ "grad_norm": 0.08296040445566177,
168
+ "learning_rate": 1.927502451102095e-05,
169
+ "loss": 0.003,
170
+ "step": 115
171
+ },
172
+ {
173
+ "epoch": 0.21937842778793418,
174
+ "grad_norm": 0.06567184627056122,
175
+ "learning_rate": 1.9150970429682316e-05,
176
+ "loss": 0.0028,
177
+ "step": 120
178
+ },
179
+ {
180
+ "epoch": 0.22851919561243145,
181
+ "grad_norm": 0.14650970697402954,
182
+ "learning_rate": 1.9017589394880515e-05,
183
+ "loss": 0.0018,
184
+ "step": 125
185
+ },
186
+ {
187
+ "epoch": 0.2376599634369287,
188
+ "grad_norm": 0.17963732779026031,
189
+ "learning_rate": 1.8875017352711547e-05,
190
+ "loss": 0.0016,
191
+ "step": 130
192
+ },
193
+ {
194
+ "epoch": 0.24680073126142596,
195
+ "grad_norm": 0.09263317286968231,
196
+ "learning_rate": 1.8723399617031754e-05,
197
+ "loss": 0.0014,
198
+ "step": 135
199
+ },
200
+ {
201
+ "epoch": 0.25594149908592323,
202
+ "grad_norm": 0.05322503671050072,
203
+ "learning_rate": 1.8562890721349434e-05,
204
+ "loss": 0.0008,
205
+ "step": 140
206
+ },
207
+ {
208
+ "epoch": 0.26508226691042047,
209
+ "grad_norm": 0.04636527970433235,
210
+ "learning_rate": 1.8393654261319504e-05,
211
+ "loss": 0.0008,
212
+ "step": 145
213
+ },
214
+ {
215
+ "epoch": 0.2742230347349177,
216
+ "grad_norm": 0.3017467260360718,
217
+ "learning_rate": 1.821586272800168e-05,
218
+ "loss": 0.0006,
219
+ "step": 150
220
+ },
221
+ {
222
+ "epoch": 0.283363802559415,
223
+ "grad_norm": 0.08996463567018509,
224
+ "learning_rate": 1.8029697332052277e-05,
225
+ "loss": 0.0006,
226
+ "step": 155
227
+ },
228
+ {
229
+ "epoch": 0.29250457038391225,
230
+ "grad_norm": 0.2036667764186859,
231
+ "learning_rate": 1.7835347819028642e-05,
232
+ "loss": 0.0006,
233
+ "step": 160
234
+ },
235
+ {
236
+ "epoch": 0.3016453382084095,
237
+ "grad_norm": 0.17612086236476898,
238
+ "learning_rate": 1.76330122759946e-05,
239
+ "loss": 0.0004,
240
+ "step": 165
241
+ },
242
+ {
243
+ "epoch": 0.31078610603290674,
244
+ "grad_norm": 0.20145903527736664,
245
+ "learning_rate": 1.7422896929623957e-05,
246
+ "loss": 0.0004,
247
+ "step": 170
248
+ },
249
+ {
250
+ "epoch": 0.31992687385740404,
251
+ "grad_norm": 0.2752271592617035,
252
+ "learning_rate": 1.720521593600787e-05,
253
+ "loss": 0.0004,
254
+ "step": 175
255
+ },
256
+ {
257
+ "epoch": 0.3290676416819013,
258
+ "grad_norm": 0.04029591754078865,
259
+ "learning_rate": 1.6980191162380298e-05,
260
+ "loss": 0.0006,
261
+ "step": 180
262
+ },
263
+ {
264
+ "epoch": 0.3382084095063985,
265
+ "grad_norm": 0.04862139746546745,
266
+ "learning_rate": 1.674805196098402e-05,
267
+ "loss": 0.0004,
268
+ "step": 185
269
+ },
270
+ {
271
+ "epoch": 0.3473491773308958,
272
+ "grad_norm": 0.0757121592760086,
273
+ "learning_rate": 1.6509034935307716e-05,
274
+ "loss": 0.0004,
275
+ "step": 190
276
+ },
277
+ {
278
+ "epoch": 0.35648994515539306,
279
+ "grad_norm": 0.13314923644065857,
280
+ "learning_rate": 1.6263383698932307e-05,
281
+ "loss": 0.0003,
282
+ "step": 195
283
+ },
284
+ {
285
+ "epoch": 0.3656307129798903,
286
+ "grad_norm": 0.28700515627861023,
287
+ "learning_rate": 1.6011348627232463e-05,
288
+ "loss": 0.0004,
289
+ "step": 200
290
+ },
291
+ {
292
+ "epoch": 0.37477148080438755,
293
+ "grad_norm": 0.5152389407157898,
294
+ "learning_rate": 1.5753186602186207e-05,
295
+ "loss": 0.0006,
296
+ "step": 205
297
+ },
298
+ {
299
+ "epoch": 0.38391224862888484,
300
+ "grad_norm": 0.25974079966545105,
301
+ "learning_rate": 1.5489160750552833e-05,
302
+ "loss": 0.0011,
303
+ "step": 210
304
+ },
305
+ {
306
+ "epoch": 0.3930530164533821,
307
+ "grad_norm": 1.2071099281311035,
308
+ "learning_rate": 1.5219540175685938e-05,
309
+ "loss": 0.0024,
310
+ "step": 215
311
+ },
312
+ {
313
+ "epoch": 0.40219378427787933,
314
+ "grad_norm": 0.17318445444107056,
315
+ "learning_rate": 1.4944599683254903e-05,
316
+ "loss": 0.0026,
317
+ "step": 220
318
+ },
319
+ {
320
+ "epoch": 0.4113345521023766,
321
+ "grad_norm": 0.20099027454853058,
322
+ "learning_rate": 1.4664619501154445e-05,
323
+ "loss": 0.0016,
324
+ "step": 225
325
+ },
326
+ {
327
+ "epoch": 0.42047531992687387,
328
+ "grad_norm": 0.25999948382377625,
329
+ "learning_rate": 1.4379884993887605e-05,
330
+ "loss": 0.0011,
331
+ "step": 230
332
+ },
333
+ {
334
+ "epoch": 0.4296160877513711,
335
+ "grad_norm": 0.043458275496959686,
336
+ "learning_rate": 1.4090686371713403e-05,
337
+ "loss": 0.0009,
338
+ "step": 235
339
+ },
340
+ {
341
+ "epoch": 0.43875685557586835,
342
+ "grad_norm": 0.3035805821418762,
343
+ "learning_rate": 1.3797318394855496e-05,
344
+ "loss": 0.0007,
345
+ "step": 240
346
+ },
347
+ {
348
+ "epoch": 0.44789762340036565,
349
+ "grad_norm": 0.37615489959716797,
350
+ "learning_rate": 1.3500080073073436e-05,
351
+ "loss": 0.0009,
352
+ "step": 245
353
+ },
354
+ {
355
+ "epoch": 0.4570383912248629,
356
+ "grad_norm": 0.04378453642129898,
357
+ "learning_rate": 1.319927436090259e-05,
358
+ "loss": 0.0004,
359
+ "step": 250
360
+ },
361
+ {
362
+ "epoch": 0.46617915904936014,
363
+ "grad_norm": 0.0630556121468544,
364
+ "learning_rate": 1.2895207848873488e-05,
365
+ "loss": 0.0006,
366
+ "step": 255
367
+ },
368
+ {
369
+ "epoch": 0.4753199268738574,
370
+ "grad_norm": 0.02288464456796646,
371
+ "learning_rate": 1.2588190451025209e-05,
372
+ "loss": 0.0002,
373
+ "step": 260
374
+ },
375
+ {
376
+ "epoch": 0.4844606946983547,
377
+ "grad_norm": 0.20390933752059937,
378
+ "learning_rate": 1.2278535089031377e-05,
379
+ "loss": 0.0004,
380
+ "step": 265
381
+ },
382
+ {
383
+ "epoch": 0.4936014625228519,
384
+ "grad_norm": 0.06073382869362831,
385
+ "learning_rate": 1.1966557373260654e-05,
386
+ "loss": 0.0002,
387
+ "step": 270
388
+ },
389
+ {
390
+ "epoch": 0.5027422303473492,
391
+ "grad_norm": 0.023382948711514473,
392
+ "learning_rate": 1.165257528109685e-05,
393
+ "loss": 0.0001,
394
+ "step": 275
395
+ },
396
+ {
397
+ "epoch": 0.5118829981718465,
398
+ "grad_norm": 0.025175806134939194,
399
+ "learning_rate": 1.1336908832846485e-05,
400
+ "loss": 0.0001,
401
+ "step": 280
402
+ },
403
+ {
404
+ "epoch": 0.5210237659963437,
405
+ "grad_norm": 0.010363736189901829,
406
+ "learning_rate": 1.1019879765564155e-05,
407
+ "loss": 0.0,
408
+ "step": 285
409
+ },
410
+ {
411
+ "epoch": 0.5301645338208409,
412
+ "grad_norm": 0.05363736301660538,
413
+ "learning_rate": 1.0701811205128115e-05,
414
+ "loss": 0.0002,
415
+ "step": 290
416
+ },
417
+ {
418
+ "epoch": 0.5393053016453382,
419
+ "grad_norm": 0.05416525900363922,
420
+ "learning_rate": 1.0383027336900356e-05,
421
+ "loss": 0.0001,
422
+ "step": 295
423
+ },
424
+ {
425
+ "epoch": 0.5484460694698354,
426
+ "grad_norm": 0.010775912553071976,
427
+ "learning_rate": 1.0063853075306792e-05,
428
+ "loss": 0.0001,
429
+ "step": 300
430
+ },
431
+ {
432
+ "epoch": 0.5575868372943327,
433
+ "grad_norm": 0.01685800403356552,
434
+ "learning_rate": 9.744613732674401e-06,
435
+ "loss": 0.0001,
436
+ "step": 305
437
+ },
438
+ {
439
+ "epoch": 0.56672760511883,
440
+ "grad_norm": 0.20173287391662598,
441
+ "learning_rate": 9.425634687662768e-06,
442
+ "loss": 0.0002,
443
+ "step": 310
444
+ },
445
+ {
446
+ "epoch": 0.5758683729433273,
447
+ "grad_norm": 0.2594170868396759,
448
+ "learning_rate": 9.107241053628058e-06,
449
+ "loss": 0.0001,
450
+ "step": 315
451
+ },
452
+ {
453
+ "epoch": 0.5850091407678245,
454
+ "grad_norm": 0.14655804634094238,
455
+ "learning_rate": 8.789757347257373e-06,
456
+ "loss": 0.0001,
457
+ "step": 320
458
+ },
459
+ {
460
+ "epoch": 0.5941499085923218,
461
+ "grad_norm": 0.1196817010641098,
462
+ "learning_rate": 8.473507157811254e-06,
463
+ "loss": 0.0002,
464
+ "step": 325
465
+ },
466
+ {
467
+ "epoch": 0.603290676416819,
468
+ "grad_norm": 0.022879011929035187,
469
+ "learning_rate": 8.158812817311474e-06,
470
+ "loss": 0.0,
471
+ "step": 330
472
+ },
473
+ {
474
+ "epoch": 0.6124314442413162,
475
+ "grad_norm": 0.04623665288090706,
476
+ "learning_rate": 7.845995072010188e-06,
477
+ "loss": 0.0003,
478
+ "step": 335
479
+ },
480
+ {
481
+ "epoch": 0.6215722120658135,
482
+ "grad_norm": 0.0164530910551548,
483
+ "learning_rate": 7.535372755475411e-06,
484
+ "loss": 0.0001,
485
+ "step": 340
486
+ },
487
+ {
488
+ "epoch": 0.6307129798903108,
489
+ "grad_norm": 0.06736422330141068,
490
+ "learning_rate": 7.22726246362592e-06,
491
+ "loss": 0.0001,
492
+ "step": 345
493
+ },
494
+ {
495
+ "epoch": 0.6398537477148081,
496
+ "grad_norm": 0.09347337484359741,
497
+ "learning_rate": 6.921978232046878e-06,
498
+ "loss": 0.0001,
499
+ "step": 350
500
+ },
501
+ {
502
+ "epoch": 0.6489945155393053,
503
+ "grad_norm": 0.0829305499792099,
504
+ "learning_rate": 6.619831215914974e-06,
505
+ "loss": 0.0001,
506
+ "step": 355
507
+ },
508
+ {
509
+ "epoch": 0.6581352833638026,
510
+ "grad_norm": 0.029442960396409035,
511
+ "learning_rate": 6.321129372859418e-06,
512
+ "loss": 0.0001,
513
+ "step": 360
514
+ },
515
+ {
516
+ "epoch": 0.6672760511882998,
517
+ "grad_norm": 0.17886511981487274,
518
+ "learning_rate": 6.026177149081949e-06,
519
+ "loss": 0.0003,
520
+ "step": 365
521
+ },
522
+ {
523
+ "epoch": 0.676416819012797,
524
+ "grad_norm": 0.004981564823538065,
525
+ "learning_rate": 5.7352751690558025e-06,
526
+ "loss": 0.0002,
527
+ "step": 370
528
+ },
529
+ {
530
+ "epoch": 0.6855575868372943,
531
+ "grad_norm": 0.09108416736125946,
532
+ "learning_rate": 5.448719929119916e-06,
533
+ "loss": 0.0002,
534
+ "step": 375
535
+ },
536
+ {
537
+ "epoch": 0.6946983546617916,
538
+ "grad_norm": 0.08255312591791153,
539
+ "learning_rate": 5.166803495280614e-06,
540
+ "loss": 0.0002,
541
+ "step": 380
542
+ },
543
+ {
544
+ "epoch": 0.7038391224862889,
545
+ "grad_norm": 0.09765861183404922,
546
+ "learning_rate": 4.889813205528895e-06,
547
+ "loss": 0.0002,
548
+ "step": 385
549
+ },
550
+ {
551
+ "epoch": 0.7129798903107861,
552
+ "grad_norm": 0.18912295997142792,
553
+ "learning_rate": 4.61803137697661e-06,
554
+ "loss": 0.0001,
555
+ "step": 390
556
+ },
557
+ {
558
+ "epoch": 0.7221206581352834,
559
+ "grad_norm": 0.03831862658262253,
560
+ "learning_rate": 4.351735018110066e-06,
561
+ "loss": 0.0001,
562
+ "step": 395
563
+ },
564
+ {
565
+ "epoch": 0.7312614259597806,
566
+ "grad_norm": 0.06283199042081833,
567
+ "learning_rate": 4.091195546454398e-06,
568
+ "loss": 0.0001,
569
+ "step": 400
570
+ },
571
+ {
572
+ "epoch": 0.7404021937842779,
573
+ "grad_norm": 0.02292688563466072,
574
+ "learning_rate": 3.8366785119363624e-06,
575
+ "loss": 0.0001,
576
+ "step": 405
577
+ },
578
+ {
579
+ "epoch": 0.7495429616087751,
580
+ "grad_norm": 0.07033329457044601,
581
+ "learning_rate": 3.5884433262276376e-06,
582
+ "loss": 0.0001,
583
+ "step": 410
584
+ },
585
+ {
586
+ "epoch": 0.7586837294332724,
587
+ "grad_norm": 0.0036680041812360287,
588
+ "learning_rate": 3.3467429983443477e-06,
589
+ "loss": 0.0002,
590
+ "step": 415
591
+ },
592
+ {
593
+ "epoch": 0.7678244972577697,
594
+ "grad_norm": 0.024505017325282097,
595
+ "learning_rate": 3.111823876772426e-06,
596
+ "loss": 0.0001,
597
+ "step": 420
598
+ },
599
+ {
600
+ "epoch": 0.7769652650822669,
601
+ "grad_norm": 0.20300976932048798,
602
+ "learning_rate": 2.883925398381585e-06,
603
+ "loss": 0.0001,
604
+ "step": 425
605
+ },
606
+ {
607
+ "epoch": 0.7861060329067642,
608
+ "grad_norm": 0.004944251384586096,
609
+ "learning_rate": 2.663279844383815e-06,
610
+ "loss": 0.0,
611
+ "step": 430
612
+ },
613
+ {
614
+ "epoch": 0.7952468007312614,
615
+ "grad_norm": 0.051807742565870285,
616
+ "learning_rate": 2.4501121035851494e-06,
617
+ "loss": 0.0001,
618
+ "step": 435
619
+ },
620
+ {
621
+ "epoch": 0.8043875685557587,
622
+ "grad_norm": 0.004597936756908894,
623
+ "learning_rate": 2.244639443172013e-06,
624
+ "loss": 0.0,
625
+ "step": 440
626
+ },
627
+ {
628
+ "epoch": 0.8135283363802559,
629
+ "grad_norm": 0.05087242275476456,
630
+ "learning_rate": 2.047071287265735e-06,
631
+ "loss": 0.0001,
632
+ "step": 445
633
+ },
634
+ {
635
+ "epoch": 0.8226691042047533,
636
+ "grad_norm": 0.09946884959936142,
637
+ "learning_rate": 1.857609003471007e-06,
638
+ "loss": 0.0001,
639
+ "step": 450
640
+ },
641
+ {
642
+ "epoch": 0.8318098720292505,
643
+ "grad_norm": 0.0069345757365226746,
644
+ "learning_rate": 1.6764456976357279e-06,
645
+ "loss": 0.0001,
646
+ "step": 455
647
+ },
648
+ {
649
+ "epoch": 0.8409506398537477,
650
+ "grad_norm": 0.0009497898281551898,
651
+ "learning_rate": 1.503766017031547e-06,
652
+ "loss": 0.0,
653
+ "step": 460
654
+ },
655
+ {
656
+ "epoch": 0.850091407678245,
657
+ "grad_norm": 0.004903269000351429,
658
+ "learning_rate": 1.339745962155613e-06,
659
+ "loss": 0.0001,
660
+ "step": 465
661
+ },
662
+ {
663
+ "epoch": 0.8592321755027422,
664
+ "grad_norm": 0.04905455559492111,
665
+ "learning_rate": 1.1845527073454045e-06,
666
+ "loss": 0.0,
667
+ "step": 470
668
+ },
669
+ {
670
+ "epoch": 0.8683729433272395,
671
+ "grad_norm": 0.01231853011995554,
672
+ "learning_rate": 1.0383444303894453e-06,
673
+ "loss": 0.0,
674
+ "step": 475
675
+ },
676
+ {
677
+ "epoch": 0.8775137111517367,
678
+ "grad_norm": 0.11394715309143066,
679
+ "learning_rate": 9.012701513075839e-07,
680
+ "loss": 0.0001,
681
+ "step": 480
682
+ },
683
+ {
684
+ "epoch": 0.886654478976234,
685
+ "grad_norm": 0.005672850646078587,
686
+ "learning_rate": 7.734695804651693e-07,
687
+ "loss": 0.0,
688
+ "step": 485
689
+ },
690
+ {
691
+ "epoch": 0.8957952468007313,
692
+ "grad_norm": 0.01644750125706196,
693
+ "learning_rate": 6.550729761758901e-07,
694
+ "loss": 0.0,
695
+ "step": 490
696
+ },
697
+ {
698
+ "epoch": 0.9049360146252285,
699
+ "grad_norm": 0.007940856739878654,
700
+ "learning_rate": 5.462010119384665e-07,
701
+ "loss": 0.0,
702
+ "step": 495
703
+ },
704
+ {
705
+ "epoch": 0.9140767824497258,
706
+ "grad_norm": 0.07022310048341751,
707
+ "learning_rate": 4.4696465344245874e-07,
708
+ "loss": 0.0001,
709
+ "step": 500
710
+ },
711
+ {
712
+ "epoch": 0.923217550274223,
713
+ "grad_norm": 0.0016659650718793273,
714
+ "learning_rate": 3.574650454685902e-07,
715
+ "loss": 0.0001,
716
+ "step": 505
717
+ },
718
+ {
719
+ "epoch": 0.9323583180987203,
720
+ "grad_norm": 0.03347240015864372,
721
+ "learning_rate": 2.777934087988532e-07,
722
+ "loss": 0.0,
723
+ "step": 510
724
+ },
725
+ {
726
+ "epoch": 0.9414990859232175,
727
+ "grad_norm": 0.004893618635833263,
728
+ "learning_rate": 2.0803094724143879e-07,
729
+ "loss": 0.0,
730
+ "step": 515
731
+ },
732
+ {
733
+ "epoch": 0.9506398537477148,
734
+ "grad_norm": 0.009828943759202957,
735
+ "learning_rate": 1.482487648653008e-07,
736
+ "loss": 0.0001,
737
+ "step": 520
738
+ },
739
+ {
740
+ "epoch": 0.9597806215722121,
741
+ "grad_norm": 0.05608784779906273,
742
+ "learning_rate": 9.85077935286749e-08,
743
+ "loss": 0.0001,
744
+ "step": 525
745
+ },
746
+ {
747
+ "epoch": 0.9689213893967094,
748
+ "grad_norm": 0.022456202656030655,
749
+ "learning_rate": 5.8858730775438465e-08,
750
+ "loss": 0.0,
751
+ "step": 530
752
+ },
753
+ {
754
+ "epoch": 0.9780621572212066,
755
+ "grad_norm": 0.02870968170464039,
756
+ "learning_rate": 2.9341988162595593e-08,
757
+ "loss": 0.0003,
758
+ "step": 535
759
+ },
760
+ {
761
+ "epoch": 0.9872029250457038,
762
+ "grad_norm": 0.1587427407503128,
763
+ "learning_rate": 9.987650071561217e-09,
764
+ "loss": 0.0,
765
+ "step": 540
766
+ },
767
+ {
768
+ "epoch": 0.9963436928702011,
769
+ "grad_norm": 0.018141670152544975,
770
+ "learning_rate": 8.154430452267381e-10,
771
+ "loss": 0.0001,
772
+ "step": 545
773
+ },
774
+ {
775
+ "epoch": 1.0,
776
+ "step": 547,
777
+ "total_flos": 1.0069004076135219e+18,
778
+ "train_loss": 0.04387825021601769,
779
+ "train_runtime": 3854.1911,
780
+ "train_samples_per_second": 18.162,
781
+ "train_steps_per_second": 0.142
782
  }
783
  ],
784
  "logging_steps": 5,
785
+ "max_steps": 547,
786
  "num_input_tokens_seen": 0,
787
  "num_train_epochs": 1,
788
  "save_steps": 500,
 
798
  "attributes": {}
799
  }
800
  },
801
+ "total_flos": 1.0069004076135219e+18,
802
  "train_batch_size": 4,
803
  "trial_name": null,
804
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aa8eff980e4f4c9f69e6d8447635378f810547bb6411bfe1a52baa3902da024c
3
- size 7160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e2d860c4dfb3438c41b1607a47974a45801b77d33bf13f55d8e4917bd7ff507
3
+ size 7096