Hugging Face's logo

chuanmew
/
aubo

Safetensors
Model card Files
xet
 Community
chuanmew commited on
Commit
a4cf48c
·
verified ·
1 Parent(s): 485fd44

test-4-checkpoint-5000

Browse files
Files changed (14) hide show
  1. test-4-checkpoint-5000/config.json +133 -0
  2. test-4-checkpoint-5000/dataset_statistics.json +253 -0
  3. test-4-checkpoint-5000/model.safetensors +3 -0
  4. test-4-checkpoint-5000/optimizer.pt +3 -0
  5. test-4-checkpoint-5000/rng_state_0.pth +3 -0
  6. test-4-checkpoint-5000/rng_state_1.pth +3 -0
  7. test-4-checkpoint-5000/rng_state_2.pth +3 -0
  8. test-4-checkpoint-5000/rng_state_3.pth +3 -0
  9. test-4-checkpoint-5000/rng_state_4.pth +3 -0
  10. test-4-checkpoint-5000/rng_state_5.pth +3 -0
  11. test-4-checkpoint-5000/rng_state_6.pth +3 -0
  12. test-4-checkpoint-5000/rng_state_7.pth +3 -0
  13. test-4-checkpoint-5000/scheduler.pt +3 -0
  14. test-4-checkpoint-5000/trainer_state.json +383 -0
test-4-checkpoint-5000/config.json ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "batch_size": 64,
3
+ "buffer_size": 64,
4
+ "from_pretrained": "oxe-g2-checkpoint-300000",
5
+ "frozen_keys": [
6
+ "*hf_model*"
7
+ ],
8
+ "model": {
9
+ "heads": {
10
+ "action": {
11
+ "args": [],
12
+ "kwargs": {
13
+ "action_dim": 7,
14
+ "action_horizon": 4,
15
+ "dropout_rate": 0.0,
16
+ "n_diffusion_samples": 1,
17
+ "readout_key": "readout_action",
18
+ "token_embedding_size": 768,
19
+ "use_map": false
20
+ },
21
+ "module": "octo.model.components.action_heads",
22
+ "name": "DiffusionActionHead"
23
+ }
24
+ },
25
+ "max_horizon": 10,
26
+ "observation_tokenizers": {
27
+ "primary": {
28
+ "args": [],
29
+ "kwargs": {
30
+ "encoder": {
31
+ "args": [],
32
+ "kwargs": {
33
+ "in_features": 6
34
+ },
35
+ "module": "octo.model.components.vit_encoders",
36
+ "name": "SmallStem16"
37
+ },
38
+ "obs_stack_keys": [
39
+ "image_primary"
40
+ ],
41
+ "task_stack_keys": [
42
+ "image_primary"
43
+ ]
44
+ },
45
+ "module": "octo.model.components.tokenizers",
46
+ "name": "ImageTokenizer"
47
+ }
48
+ },
49
+ "readouts": {
50
+ "action": 1
51
+ },
52
+ "repeat_task_tokens": true,
53
+ "task_tokenizers": {
54
+ "language": {
55
+ "args": [],
56
+ "kwargs": {
57
+ "encoder": "t5-base",
58
+ "finetune_encoder": false
59
+ },
60
+ "module": "octo.model.components.tokenizers",
61
+ "name": "LanguageTokenizer"
62
+ }
63
+ },
64
+ "token_embedding_size": 768,
65
+ "transformer_kwargs": {
66
+ "add_position_embedding": false,
67
+ "attention_dropout_rate": 0.0,
68
+ "dropout_rate": 0.0,
69
+ "mlp_dim": 3072,
70
+ "num_attention_heads": 12,
71
+ "num_layers": 12
72
+ },
73
+ "use_correct_attention": true
74
+ },
75
+ "normalize_method": "sign",
76
+ "obs_token_nums": {
77
+ "primary": 256
78
+ },
79
+ "sampler_num_samples": 2560000,
80
+ "seed": 2,
81
+ "subsample_length": 99999,
82
+ "text_processor": {
83
+ "args": [],
84
+ "kwargs": {
85
+ "encode_with_model": false,
86
+ "tokenizer_kwargs": {
87
+ "max_length": 16,
88
+ "padding": "max_length",
89
+ "return_tensors": "np",
90
+ "truncation": true
91
+ },
92
+ "tokenizer_name": "t5-base"
93
+ },
94
+ "module": "octo.components.text_processing",
95
+ "name": "HFTokenizer"
96
+ },
97
+ "training_arguments": {
98
+ "bf16": true,
99
+ "dataloader_pin_memory": true,
100
+ "ddp_find_unused_parameters": true,
101
+ "eval_steps": 5000,
102
+ "gradient_accumulation_steps": 4,
103
+ "learning_rate": 5e-05,
104
+ "logging_nan_inf_filter": false,
105
+ "logging_steps": 100,
106
+ "lr_scheduler_kwargs": {
107
+ "decay_type": "cosine",
108
+ "min_lr_ratio": 0.05,
109
+ "num_decay_steps": 298000
110
+ },
111
+ "lr_scheduler_type": "constant",
112
+ "max_grad_norm": 1.0,
113
+ "max_steps": 300000,
114
+ "optim": "adamw_torch_fused",
115
+ "per_device_eval_batch_size": 1,
116
+ "per_device_train_batch_size": 1,
117
+ "report_to": "wandb",
118
+ "save_safetensors": true,
119
+ "save_steps": 500,
120
+ "torch_compile": true,
121
+ "warmup_steps": 2000,
122
+ "weight_decay": 0.1
123
+ },
124
+ "training_keys": [
125
+ "*action*",
126
+ "*task_projections*",
127
+ "*obs_projections*",
128
+ "*readout_embeddings*",
129
+ "*task_pos_embeddings*",
130
+ "*obs_pos_embeddings*",
131
+ "*observation_tokenizers*"
132
+ ]
133
+ }
test-4-checkpoint-5000/dataset_statistics.json ADDED
@@ -0,0 +1,253 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "action": {
4
+ "min": [
5
+ -0.009999999776482582,
6
+ -0.009999999776482582,
7
+ -0.009999999776482582,
8
+ -0.06283185631036758,
9
+ -0.06283185631036758,
10
+ -0.06283185631036758,
11
+ 0.0
12
+ ],
13
+ "max": [
14
+ 0.009999999776482582,
15
+ 0.009999999776482582,
16
+ 0.009999999776482582,
17
+ 0.06283185631036758,
18
+ 0.06283185631036758,
19
+ 0.06283185631036758,
20
+ 1.0
21
+ ],
22
+ "mean": [
23
+ -0.0010779668760702325,
24
+ -0.000870422678087375,
25
+ -0.0009524932320082003,
26
+ -0.00011404143052376722,
27
+ 0.001646163140721359,
28
+ 0.0006842485724881945,
29
+ 0.14094065714126566
30
+ ],
31
+ "std": [
32
+ 0.0050928476793778805,
33
+ 0.0052109778002382215,
34
+ 0.0062727072406226275,
35
+ 0.01305372613557191,
36
+ 0.014969692486291531,
37
+ 0.021519857205295197,
38
+ 0.3479603303557627
39
+ ],
40
+ "count": [
41
+ 12672
42
+ ],
43
+ "mask": [
44
+ true,
45
+ true,
46
+ true,
47
+ true,
48
+ true,
49
+ true,
50
+ false
51
+ ]
52
+ },
53
+ "index": {
54
+ "min": [
55
+ 0
56
+ ],
57
+ "max": [
58
+ 12671
59
+ ],
60
+ "mean": [
61
+ 6335.5
62
+ ],
63
+ "std": [
64
+ 3658.0912941951938
65
+ ],
66
+ "count": [
67
+ 12672
68
+ ]
69
+ },
70
+ "frame_index": {
71
+ "min": [
72
+ 0
73
+ ],
74
+ "max": [
75
+ 225
76
+ ],
77
+ "mean": [
78
+ 55.168955176767675
79
+ ],
80
+ "std": [
81
+ 39.5185866669507
82
+ ],
83
+ "count": [
84
+ 12672
85
+ ]
86
+ },
87
+ "observation.images.main_camera": {
88
+ "min": [
89
+ [
90
+ [
91
+ 0.0
92
+ ]
93
+ ],
94
+ [
95
+ [
96
+ 0.0
97
+ ]
98
+ ],
99
+ [
100
+ [
101
+ 0.0
102
+ ]
103
+ ]
104
+ ],
105
+ "max": [
106
+ [
107
+ [
108
+ 1.0
109
+ ]
110
+ ],
111
+ [
112
+ [
113
+ 1.0
114
+ ]
115
+ ],
116
+ [
117
+ [
118
+ 1.0
119
+ ]
120
+ ]
121
+ ],
122
+ "mean": [
123
+ [
124
+ [
125
+ 0.56009308397514
126
+ ]
127
+ ],
128
+ [
129
+ [
130
+ 0.5088963212208076
131
+ ]
132
+ ],
133
+ [
134
+ [
135
+ 0.47626117961949455
136
+ ]
137
+ ]
138
+ ],
139
+ "std": [
140
+ [
141
+ [
142
+ 0.27688289864943944
143
+ ]
144
+ ],
145
+ [
146
+ [
147
+ 0.26883972589038224
148
+ ]
149
+ ],
150
+ [
151
+ [
152
+ 0.29031452356480364
153
+ ]
154
+ ]
155
+ ],
156
+ "count": [
157
+ 11006
158
+ ]
159
+ },
160
+ "timestamp": {
161
+ "min": [
162
+ 0.0
163
+ ],
164
+ "max": [
165
+ 22.5
166
+ ],
167
+ "mean": [
168
+ 5.516895517676768
169
+ ],
170
+ "std": [
171
+ 3.9518586666950704
172
+ ],
173
+ "count": [
174
+ 12672
175
+ ]
176
+ },
177
+ "task_index": {
178
+ "min": [
179
+ 0
180
+ ],
181
+ "max": [
182
+ 3
183
+ ],
184
+ "mean": [
185
+ 1.6840277777777777
186
+ ],
187
+ "std": [
188
+ 1.1113384592218372
189
+ ],
190
+ "count": [
191
+ 12672
192
+ ]
193
+ },
194
+ "episode_index": {
195
+ "min": [
196
+ 0
197
+ ],
198
+ "max": [
199
+ 128
200
+ ],
201
+ "mean": [
202
+ 63.14946338383838
203
+ ],
204
+ "std": [
205
+ 35.1467891774593
206
+ ],
207
+ "count": [
208
+ 12672
209
+ ]
210
+ },
211
+ "observation.state": {
212
+ "min": [
213
+ -0.6913691163063049,
214
+ -0.25204867124557495,
215
+ -0.22021731734275818,
216
+ -3.140528440475464,
217
+ -1.213579535484314,
218
+ -3.1340115070343018,
219
+ 0.0
220
+ ],
221
+ "max": [
222
+ -0.26134854555130005,
223
+ 0.40225377678871155,
224
+ 0.4610653519630432,
225
+ 3.1412761211395264,
226
+ 0.644866406917572,
227
+ 3.1401240825653076,
228
+ 1.0
229
+ ],
230
+ "mean": [
231
+ -0.4765559839921994,
232
+ 0.03277433536791847,
233
+ 0.01813845893630427,
234
+ 1.4840563985447381,
235
+ -0.1872646863000659,
236
+ -1.201557804914303,
237
+ 0.13328598546378098
238
+ ],
239
+ "std": [
240
+ 0.09873974979292631,
241
+ 0.12291375325367411,
242
+ 0.13182701724212956,
243
+ 2.530463134923631,
244
+ 0.39532953579234853,
245
+ 1.3639962652103377,
246
+ 0.33988356830665484
247
+ ],
248
+ "count": [
249
+ 12672
250
+ ]
251
+ }
252
+ }
253
+ ]
test-4-checkpoint-5000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ae6ac68cb9b461e1f89bdd9b01b1256916a80a136807a262651f3b59a6d23b9
3
+ size 401078544
test-4-checkpoint-5000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8598362dded9dd4c300db6de6559e5968093ee3760031625236df52e3dd09ff
3
+ size 26356666
test-4-checkpoint-5000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46db914519797fdc01e13e22273f1a0fc5d32bdcc34f7b7fc3278bb4ea933e48
3
+ size 15984
test-4-checkpoint-5000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1fb84ef196c00bfc9155a735be9779af5ede71f91af79fb144a2f83627869f0
3
+ size 15984
test-4-checkpoint-5000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c18f634bdd9bad79d75a06d58bdbcd3c140751f3be74c46b7d5a51fd8fbb874a
3
+ size 15984
test-4-checkpoint-5000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:257fd12da17d5a02f89966cf943fdda9b51c218393f7c9a79dbdf33334d144c3
3
+ size 15984
test-4-checkpoint-5000/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d1da8125d45758ca75e1c0a799a0c1e7e3ed5aed922671a05ed532f8ae9eb81
3
+ size 15984
test-4-checkpoint-5000/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:212cb7218b5f30fa0eef9b782a0070f10e61b523e8f13e01f8f090465f47da3c
3
+ size 15984
test-4-checkpoint-5000/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:436805e8649c608e2b99d6c714c009dbb376e292d2c5cb286c3f93574c09f88e
3
+ size 15984
test-4-checkpoint-5000/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e90d86be13c1c760f80327d41f993d059911d8333e49a3c254ec83b498d35cee
3
+ size 15984
test-4-checkpoint-5000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5efa72feef3a398c4152d1d39792b540dbc6e38d02d8ab41bc4a583b249405d
3
+ size 1064
test-4-checkpoint-5000/trainer_state.json ADDED
@@ -0,0 +1,383 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "eval_steps": 5000,
6
+ "global_step": 5000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.08,
13
+ "grad_norm": 3.25,
14
+ "learning_rate": 5e-05,
15
+ "loss": 0.5172,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.16,
20
+ "grad_norm": 3.921875,
21
+ "learning_rate": 5e-05,
22
+ "loss": 0.3445,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 0.24,
27
+ "grad_norm": 4.09375,
28
+ "learning_rate": 5e-05,
29
+ "loss": 0.3051,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 0.32,
34
+ "grad_norm": 3.09375,
35
+ "learning_rate": 5e-05,
36
+ "loss": 0.2824,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 0.4,
41
+ "grad_norm": 3.09375,
42
+ "learning_rate": 5e-05,
43
+ "loss": 0.265,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 0.48,
48
+ "grad_norm": 2.484375,
49
+ "learning_rate": 5e-05,
50
+ "loss": 0.2517,
51
+ "step": 600
52
+ },
53
+ {
54
+ "epoch": 0.56,
55
+ "grad_norm": 3.109375,
56
+ "learning_rate": 5e-05,
57
+ "loss": 0.2448,
58
+ "step": 700
59
+ },
60
+ {
61
+ "epoch": 0.64,
62
+ "grad_norm": 2.5,
63
+ "learning_rate": 5e-05,
64
+ "loss": 0.2374,
65
+ "step": 800
66
+ },
67
+ {
68
+ "epoch": 0.72,
69
+ "grad_norm": 2.0625,
70
+ "learning_rate": 5e-05,
71
+ "loss": 0.2327,
72
+ "step": 900
73
+ },
74
+ {
75
+ "epoch": 0.8,
76
+ "grad_norm": 2.765625,
77
+ "learning_rate": 5e-05,
78
+ "loss": 0.2291,
79
+ "step": 1000
80
+ },
81
+ {
82
+ "epoch": 0.88,
83
+ "grad_norm": 2.125,
84
+ "learning_rate": 5e-05,
85
+ "loss": 0.2243,
86
+ "step": 1100
87
+ },
88
+ {
89
+ "epoch": 0.96,
90
+ "grad_norm": 3.03125,
91
+ "learning_rate": 5e-05,
92
+ "loss": 0.2209,
93
+ "step": 1200
94
+ },
95
+ {
96
+ "epoch": 1.04,
97
+ "grad_norm": 2.265625,
98
+ "learning_rate": 5e-05,
99
+ "loss": 0.2167,
100
+ "step": 1300
101
+ },
102
+ {
103
+ "epoch": 1.12,
104
+ "grad_norm": 2.546875,
105
+ "learning_rate": 5e-05,
106
+ "loss": 0.2142,
107
+ "step": 1400
108
+ },
109
+ {
110
+ "epoch": 1.2,
111
+ "grad_norm": 2.171875,
112
+ "learning_rate": 5e-05,
113
+ "loss": 0.2124,
114
+ "step": 1500
115
+ },
116
+ {
117
+ "epoch": 1.28,
118
+ "grad_norm": 2.609375,
119
+ "learning_rate": 5e-05,
120
+ "loss": 0.2116,
121
+ "step": 1600
122
+ },
123
+ {
124
+ "epoch": 1.3599999999999999,
125
+ "grad_norm": 2.078125,
126
+ "learning_rate": 5e-05,
127
+ "loss": 0.2097,
128
+ "step": 1700
129
+ },
130
+ {
131
+ "epoch": 1.44,
132
+ "grad_norm": 2.296875,
133
+ "learning_rate": 5e-05,
134
+ "loss": 0.2069,
135
+ "step": 1800
136
+ },
137
+ {
138
+ "epoch": 1.52,
139
+ "grad_norm": 2.1875,
140
+ "learning_rate": 5e-05,
141
+ "loss": 0.2037,
142
+ "step": 1900
143
+ },
144
+ {
145
+ "epoch": 1.6,
146
+ "grad_norm": 2.21875,
147
+ "learning_rate": 5e-05,
148
+ "loss": 0.2025,
149
+ "step": 2000
150
+ },
151
+ {
152
+ "epoch": 1.6800000000000002,
153
+ "grad_norm": 2.640625,
154
+ "learning_rate": 5e-05,
155
+ "loss": 0.2019,
156
+ "step": 2100
157
+ },
158
+ {
159
+ "epoch": 1.76,
160
+ "grad_norm": 3.015625,
161
+ "learning_rate": 5e-05,
162
+ "loss": 0.1997,
163
+ "step": 2200
164
+ },
165
+ {
166
+ "epoch": 1.8399999999999999,
167
+ "grad_norm": 2.515625,
168
+ "learning_rate": 5e-05,
169
+ "loss": 0.1987,
170
+ "step": 2300
171
+ },
172
+ {
173
+ "epoch": 1.92,
174
+ "grad_norm": 2.390625,
175
+ "learning_rate": 5e-05,
176
+ "loss": 0.1974,
177
+ "step": 2400
178
+ },
179
+ {
180
+ "epoch": 2.0,
181
+ "grad_norm": 1.9921875,
182
+ "learning_rate": 5e-05,
183
+ "loss": 0.1978,
184
+ "step": 2500
185
+ },
186
+ {
187
+ "epoch": 2.08,
188
+ "grad_norm": 2.4375,
189
+ "learning_rate": 5e-05,
190
+ "loss": 0.1949,
191
+ "step": 2600
192
+ },
193
+ {
194
+ "epoch": 2.16,
195
+ "grad_norm": 2.453125,
196
+ "learning_rate": 5e-05,
197
+ "loss": 0.1928,
198
+ "step": 2700
199
+ },
200
+ {
201
+ "epoch": 2.24,
202
+ "grad_norm": 3.1875,
203
+ "learning_rate": 5e-05,
204
+ "loss": 0.1935,
205
+ "step": 2800
206
+ },
207
+ {
208
+ "epoch": 2.32,
209
+ "grad_norm": 2.078125,
210
+ "learning_rate": 5e-05,
211
+ "loss": 0.1912,
212
+ "step": 2900
213
+ },
214
+ {
215
+ "epoch": 2.4,
216
+ "grad_norm": 2.0,
217
+ "learning_rate": 5e-05,
218
+ "loss": 0.1898,
219
+ "step": 3000
220
+ },
221
+ {
222
+ "epoch": 2.48,
223
+ "grad_norm": 2.234375,
224
+ "learning_rate": 5e-05,
225
+ "loss": 0.1881,
226
+ "step": 3100
227
+ },
228
+ {
229
+ "epoch": 2.56,
230
+ "grad_norm": 2.4375,
231
+ "learning_rate": 5e-05,
232
+ "loss": 0.1862,
233
+ "step": 3200
234
+ },
235
+ {
236
+ "epoch": 2.64,
237
+ "grad_norm": 1.9921875,
238
+ "learning_rate": 5e-05,
239
+ "loss": 0.1871,
240
+ "step": 3300
241
+ },
242
+ {
243
+ "epoch": 2.7199999999999998,
244
+ "grad_norm": 2.1875,
245
+ "learning_rate": 5e-05,
246
+ "loss": 0.1853,
247
+ "step": 3400
248
+ },
249
+ {
250
+ "epoch": 2.8,
251
+ "grad_norm": 2.34375,
252
+ "learning_rate": 5e-05,
253
+ "loss": 0.185,
254
+ "step": 3500
255
+ },
256
+ {
257
+ "epoch": 2.88,
258
+ "grad_norm": 2.078125,
259
+ "learning_rate": 5e-05,
260
+ "loss": 0.1853,
261
+ "step": 3600
262
+ },
263
+ {
264
+ "epoch": 2.96,
265
+ "grad_norm": 2.015625,
266
+ "learning_rate": 5e-05,
267
+ "loss": 0.1849,
268
+ "step": 3700
269
+ },
270
+ {
271
+ "epoch": 3.04,
272
+ "grad_norm": 2.8125,
273
+ "learning_rate": 5e-05,
274
+ "loss": 0.1846,
275
+ "step": 3800
276
+ },
277
+ {
278
+ "epoch": 3.12,
279
+ "grad_norm": 2.40625,
280
+ "learning_rate": 5e-05,
281
+ "loss": 0.1833,
282
+ "step": 3900
283
+ },
284
+ {
285
+ "epoch": 3.2,
286
+ "grad_norm": 2.140625,
287
+ "learning_rate": 5e-05,
288
+ "loss": 0.1828,
289
+ "step": 4000
290
+ },
291
+ {
292
+ "epoch": 3.2800000000000002,
293
+ "grad_norm": 2.140625,
294
+ "learning_rate": 5e-05,
295
+ "loss": 0.1821,
296
+ "step": 4100
297
+ },
298
+ {
299
+ "epoch": 3.36,
300
+ "grad_norm": 2.203125,
301
+ "learning_rate": 5e-05,
302
+ "loss": 0.1816,
303
+ "step": 4200
304
+ },
305
+ {
306
+ "epoch": 3.44,
307
+ "grad_norm": 2.515625,
308
+ "learning_rate": 5e-05,
309
+ "loss": 0.181,
310
+ "step": 4300
311
+ },
312
+ {
313
+ "epoch": 3.52,
314
+ "grad_norm": 1.84375,
315
+ "learning_rate": 5e-05,
316
+ "loss": 0.1803,
317
+ "step": 4400
318
+ },
319
+ {
320
+ "epoch": 3.6,
321
+ "grad_norm": 2.046875,
322
+ "learning_rate": 5e-05,
323
+ "loss": 0.1798,
324
+ "step": 4500
325
+ },
326
+ {
327
+ "epoch": 3.68,
328
+ "grad_norm": 1.9296875,
329
+ "learning_rate": 5e-05,
330
+ "loss": 0.1778,
331
+ "step": 4600
332
+ },
333
+ {
334
+ "epoch": 3.76,
335
+ "grad_norm": 2.421875,
336
+ "learning_rate": 5e-05,
337
+ "loss": 0.1782,
338
+ "step": 4700
339
+ },
340
+ {
341
+ "epoch": 3.84,
342
+ "grad_norm": 2.25,
343
+ "learning_rate": 5e-05,
344
+ "loss": 0.1755,
345
+ "step": 4800
346
+ },
347
+ {
348
+ "epoch": 3.92,
349
+ "grad_norm": 2.296875,
350
+ "learning_rate": 5e-05,
351
+ "loss": 0.1771,
352
+ "step": 4900
353
+ },
354
+ {
355
+ "epoch": 4.0,
356
+ "grad_norm": 1.8203125,
357
+ "learning_rate": 5e-05,
358
+ "loss": 0.1751,
359
+ "step": 5000
360
+ }
361
+ ],
362
+ "logging_steps": 100,
363
+ "max_steps": 300000,
364
+ "num_input_tokens_seen": 0,
365
+ "num_train_epochs": 240,
366
+ "save_steps": 500,
367
+ "stateful_callbacks": {
368
+ "TrainerControl": {
369
+ "args": {
370
+ "should_epoch_stop": false,
371
+ "should_evaluate": false,
372
+ "should_log": false,
373
+ "should_save": true,
374
+ "should_training_stop": false
375
+ },
376
+ "attributes": {}
377
+ }
378
+ },
379
+ "total_flos": 0.0,
380
+ "train_batch_size": 1,
381
+ "trial_name": null,
382
+ "trial_params": null
383
+ }