Hugging Face's logo

rshrott
/
ryan_model314

Image Classification
Transformers
TensorBoard
Safetensors
vit
Generated from Trainer
Model card Files
xet
Metrics Community
rshrott commited on
Commit
07f4476
·
verified ·
1 Parent(s): 27f95d6

🍻 cheers

Browse files
Files changed (6) hide show
  1. README.md +5 -4
  2. all_results.json +10 -10
  3. eval_results.json +6 -6
  4. runs/Mar26_19-55-16_ryanserver/events.out.tfevents.1711499103.ryanserver.31878.1 +3 -0
  5. train_results.json +4 -4
  6. trainer_state.json +925 -125
README.md CHANGED
@@ -2,6 +2,7 @@
2
  license: apache-2.0
3
  base_model: google/vit-base-patch16-224-in21k
4
  tags:
 
5
  - generated_from_trainer
6
  model-index:
7
  - name: ryan_model314
@@ -13,11 +14,11 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  # ryan_model314
15
 
16
- This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.2670
19
- - Na Accuracy: 0.962
20
- - Ordinal Accuracy: 0.6760
21
 
22
  ## Model description
23
 
 
2
  license: apache-2.0
3
  base_model: google/vit-base-patch16-224-in21k
4
  tags:
5
+ - image-classification
6
  - generated_from_trainer
7
  model-index:
8
  - name: ryan_model314
 
14
 
15
  # ryan_model314
16
 
17
+ This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.2404
20
+ - Na Accuracy: 0.96
21
+ - Ordinal Accuracy: 0.5921
22
 
23
  ## Model description
24
 
all_results.json CHANGED
@@ -1,14 +1,14 @@
1
  {
2
  "epoch": 4.0,
3
  "eval_accuracy": 0.628,
4
- "eval_loss": 0.29386183619499207,
5
- "eval_na_accuracy": 0.945,
6
- "eval_ordinal_accuracy": 0.5705521472392638,
7
- "eval_runtime": 8.2243,
8
- "eval_samples_per_second": 24.318,
9
- "eval_steps_per_second": 3.04,
10
- "train_loss": 0.212149089468377,
11
- "train_runtime": 378.6413,
12
- "train_samples_per_second": 10.564,
13
- "train_steps_per_second": 0.666
14
  }
 
1
  {
2
  "epoch": 4.0,
3
  "eval_accuracy": 0.628,
4
+ "eval_loss": 0.24043463170528412,
5
+ "eval_na_accuracy": 0.96,
6
+ "eval_ordinal_accuracy": 0.5920745920745921,
7
+ "eval_runtime": 19.2933,
8
+ "eval_samples_per_second": 25.916,
9
+ "eval_steps_per_second": 3.265,
10
+ "train_loss": 0.15608444792060808,
11
+ "train_runtime": 1745.0942,
12
+ "train_samples_per_second": 11.461,
13
+ "train_steps_per_second": 0.717
14
  }
eval_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "epoch": 4.0,
3
- "eval_loss": 0.29386183619499207,
4
- "eval_na_accuracy": 0.945,
5
- "eval_ordinal_accuracy": 0.5705521472392638,
6
- "eval_runtime": 8.2243,
7
- "eval_samples_per_second": 24.318,
8
- "eval_steps_per_second": 3.04
9
  }
 
1
  {
2
  "epoch": 4.0,
3
+ "eval_loss": 0.24043463170528412,
4
+ "eval_na_accuracy": 0.96,
5
+ "eval_ordinal_accuracy": 0.5920745920745921,
6
+ "eval_runtime": 19.2933,
7
+ "eval_samples_per_second": 25.916,
8
+ "eval_steps_per_second": 3.265
9
  }
runs/Mar26_19-55-16_ryanserver/events.out.tfevents.1711499103.ryanserver.31878.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2880e2c2215eaef570c0be996f15cabceae3651e279b2ce21eac8443260ed04c
3
+ size 474
train_results.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "epoch": 4.0,
3
- "train_loss": 0.212149089468377,
4
- "train_runtime": 378.6413,
5
- "train_samples_per_second": 10.564,
6
- "train_steps_per_second": 0.666
7
  }
 
1
  {
2
  "epoch": 4.0,
3
+ "train_loss": 0.15608444792060808,
4
+ "train_runtime": 1745.0942,
5
+ "train_samples_per_second": 11.461,
6
+ "train_steps_per_second": 0.717
7
  }
trainer_state.json CHANGED
@@ -1,224 +1,1024 @@
1
  {
2
- "best_metric": 0.29386183619499207,
3
- "best_model_checkpoint": "./ryan_model314/checkpoint-200",
4
  "epoch": 4.0,
5
  "eval_steps": 100,
6
- "global_step": 252,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.16,
13
- "grad_norm": 1.1858803033828735,
14
- "learning_rate": 0.00019206349206349208,
15
- "loss": 0.5736,
16
  "step": 10
17
  },
18
  {
19
- "epoch": 0.32,
20
- "grad_norm": 1.41181480884552,
21
- "learning_rate": 0.00018412698412698412,
22
- "loss": 0.4142,
23
  "step": 20
24
  },
25
  {
26
- "epoch": 0.48,
27
- "grad_norm": 0.6347964406013489,
28
- "learning_rate": 0.0001761904761904762,
29
- "loss": 0.3916,
30
  "step": 30
31
  },
32
  {
33
- "epoch": 0.63,
34
- "grad_norm": 0.7893273234367371,
35
- "learning_rate": 0.00016825396825396826,
36
- "loss": 0.3628,
37
  "step": 40
38
  },
39
  {
40
- "epoch": 0.79,
41
- "grad_norm": 0.9887136816978455,
42
- "learning_rate": 0.00016031746031746033,
43
- "loss": 0.3852,
44
  "step": 50
45
  },
46
  {
47
- "epoch": 0.95,
48
- "grad_norm": 1.1596781015396118,
49
- "learning_rate": 0.00015238095238095237,
50
- "loss": 0.3962,
51
  "step": 60
52
  },
53
  {
54
- "epoch": 1.11,
55
- "grad_norm": 1.1897984743118286,
56
- "learning_rate": 0.00014444444444444444,
57
- "loss": 0.2923,
58
  "step": 70
59
  },
60
  {
61
- "epoch": 1.27,
62
- "grad_norm": 0.5242781639099121,
63
- "learning_rate": 0.0001365079365079365,
64
- "loss": 0.2335,
65
  "step": 80
66
  },
67
  {
68
- "epoch": 1.43,
69
- "grad_norm": 1.0704305171966553,
70
- "learning_rate": 0.00012857142857142858,
71
- "loss": 0.268,
72
  "step": 90
73
  },
74
  {
75
- "epoch": 1.59,
76
- "grad_norm": 0.852606475353241,
77
- "learning_rate": 0.00012063492063492063,
78
- "loss": 0.1969,
79
  "step": 100
80
  },
81
  {
82
- "epoch": 1.59,
83
- "eval_loss": 0.2954840064048767,
84
- "eval_na_accuracy": 0.945,
85
- "eval_ordinal_accuracy": 0.4785276073619632,
86
- "eval_runtime": 26.5841,
87
- "eval_samples_per_second": 7.523,
88
- "eval_steps_per_second": 0.94,
89
  "step": 100
90
  },
91
  {
92
- "epoch": 1.75,
93
- "grad_norm": 0.6733121871948242,
94
- "learning_rate": 0.0001126984126984127,
95
- "loss": 0.2328,
96
  "step": 110
97
  },
98
  {
99
- "epoch": 1.9,
100
- "grad_norm": 0.9004744291305542,
101
- "learning_rate": 0.00010476190476190477,
102
- "loss": 0.2248,
103
  "step": 120
104
  },
105
  {
106
- "epoch": 2.06,
107
- "grad_norm": 3.1687183380126953,
108
- "learning_rate": 9.682539682539682e-05,
109
- "loss": 0.1717,
110
  "step": 130
111
  },
112
  {
113
- "epoch": 2.22,
114
- "grad_norm": 0.6659616827964783,
115
- "learning_rate": 8.888888888888889e-05,
116
- "loss": 0.145,
117
  "step": 140
118
  },
119
  {
120
- "epoch": 2.38,
121
- "grad_norm": 0.6846858859062195,
122
- "learning_rate": 8.095238095238096e-05,
123
- "loss": 0.1157,
124
  "step": 150
125
  },
126
  {
127
- "epoch": 2.54,
128
- "grad_norm": 0.6155730485916138,
129
- "learning_rate": 7.301587301587302e-05,
130
- "loss": 0.1424,
131
  "step": 160
132
  },
133
  {
134
- "epoch": 2.7,
135
- "grad_norm": 0.6559838056564331,
136
- "learning_rate": 6.507936507936509e-05,
137
- "loss": 0.1041,
138
  "step": 170
139
  },
140
  {
141
- "epoch": 2.86,
142
- "grad_norm": 0.4849882423877716,
143
- "learning_rate": 5.714285714285714e-05,
144
- "loss": 0.1339,
145
  "step": 180
146
  },
147
  {
148
- "epoch": 3.02,
149
- "grad_norm": 0.6311644911766052,
150
- "learning_rate": 4.9206349206349204e-05,
151
- "loss": 0.103,
152
  "step": 190
153
  },
154
  {
155
- "epoch": 3.17,
156
- "grad_norm": 0.8323171138763428,
157
- "learning_rate": 4.126984126984127e-05,
158
- "loss": 0.0746,
159
  "step": 200
160
  },
161
  {
162
- "epoch": 3.17,
163
- "eval_loss": 0.29386183619499207,
164
- "eval_na_accuracy": 0.945,
165
- "eval_ordinal_accuracy": 0.5705521472392638,
166
- "eval_runtime": 8.3287,
167
- "eval_samples_per_second": 24.013,
168
- "eval_steps_per_second": 3.002,
169
  "step": 200
170
  },
171
  {
172
- "epoch": 3.33,
173
- "grad_norm": 0.8798254132270813,
174
- "learning_rate": 3.3333333333333335e-05,
175
- "loss": 0.086,
176
  "step": 210
177
  },
178
  {
179
- "epoch": 3.49,
180
- "grad_norm": 0.37054240703582764,
181
- "learning_rate": 2.5396825396825397e-05,
182
- "loss": 0.0933,
183
  "step": 220
184
  },
185
  {
186
- "epoch": 3.65,
187
- "grad_norm": 0.46293869614601135,
188
- "learning_rate": 1.746031746031746e-05,
189
- "loss": 0.066,
190
  "step": 230
191
  },
192
  {
193
- "epoch": 3.81,
194
- "grad_norm": 0.3859086334705353,
195
- "learning_rate": 9.523809523809523e-06,
196
- "loss": 0.0649,
197
  "step": 240
198
  },
199
  {
200
- "epoch": 3.97,
201
- "grad_norm": 0.300207257270813,
202
- "learning_rate": 1.5873015873015873e-06,
203
- "loss": 0.0623,
204
  "step": 250
205
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
206
  {
207
  "epoch": 4.0,
208
- "step": 252,
209
- "total_flos": 3.0997907103744e+17,
210
- "train_loss": 0.212149089468377,
211
- "train_runtime": 378.6413,
212
- "train_samples_per_second": 10.564,
213
- "train_steps_per_second": 0.666
214
  }
215
  ],
216
  "logging_steps": 10,
217
- "max_steps": 252,
218
  "num_input_tokens_seen": 0,
219
  "num_train_epochs": 4,
220
  "save_steps": 100,
221
- "total_flos": 3.0997907103744e+17,
222
  "train_batch_size": 16,
223
  "trial_name": null,
224
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.24043463170528412,
3
+ "best_model_checkpoint": "./ryan_model314/checkpoint-500",
4
  "epoch": 4.0,
5
  "eval_steps": 100,
6
+ "global_step": 1252,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.03,
13
+ "grad_norm": 0.962435245513916,
14
+ "learning_rate": 0.00019840255591054313,
15
+ "loss": 0.5392,
16
  "step": 10
17
  },
18
  {
19
+ "epoch": 0.06,
20
+ "grad_norm": 1.411149024963379,
21
+ "learning_rate": 0.00019680511182108628,
22
+ "loss": 0.3886,
23
  "step": 20
24
  },
25
  {
26
+ "epoch": 0.1,
27
+ "grad_norm": 0.624254584312439,
28
+ "learning_rate": 0.0001952076677316294,
29
+ "loss": 0.3602,
30
  "step": 30
31
  },
32
  {
33
+ "epoch": 0.13,
34
+ "grad_norm": 1.0637298822402954,
35
+ "learning_rate": 0.00019361022364217253,
36
+ "loss": 0.3841,
37
  "step": 40
38
  },
39
  {
40
+ "epoch": 0.16,
41
+ "grad_norm": 1.1236467361450195,
42
+ "learning_rate": 0.00019201277955271565,
43
+ "loss": 0.3312,
44
  "step": 50
45
  },
46
  {
47
+ "epoch": 0.19,
48
+ "grad_norm": 0.6737650632858276,
49
+ "learning_rate": 0.0001904153354632588,
50
+ "loss": 0.3419,
51
  "step": 60
52
  },
53
  {
54
+ "epoch": 0.22,
55
+ "grad_norm": 1.0528877973556519,
56
+ "learning_rate": 0.00018881789137380192,
57
+ "loss": 0.348,
58
  "step": 70
59
  },
60
  {
61
+ "epoch": 0.26,
62
+ "grad_norm": 1.2546306848526,
63
+ "learning_rate": 0.00018722044728434505,
64
+ "loss": 0.3884,
65
  "step": 80
66
  },
67
  {
68
+ "epoch": 0.29,
69
+ "grad_norm": 1.552256464958191,
70
+ "learning_rate": 0.0001856230031948882,
71
+ "loss": 0.4075,
72
  "step": 90
73
  },
74
  {
75
+ "epoch": 0.32,
76
+ "grad_norm": 1.334892988204956,
77
+ "learning_rate": 0.00018402555910543132,
78
+ "loss": 0.3682,
79
  "step": 100
80
  },
81
  {
82
+ "epoch": 0.32,
83
+ "eval_loss": 0.3208509385585785,
84
+ "eval_na_accuracy": 0.942,
85
+ "eval_ordinal_accuracy": 0.5268065268065268,
86
+ "eval_runtime": 54.8874,
87
+ "eval_samples_per_second": 9.11,
88
+ "eval_steps_per_second": 1.148,
89
  "step": 100
90
  },
91
  {
92
+ "epoch": 0.35,
93
+ "grad_norm": 1.1707366704940796,
94
+ "learning_rate": 0.00018242811501597444,
95
+ "loss": 0.274,
96
  "step": 110
97
  },
98
  {
99
+ "epoch": 0.38,
100
+ "grad_norm": 0.9099497199058533,
101
+ "learning_rate": 0.00018083067092651756,
102
+ "loss": 0.3335,
103
  "step": 120
104
  },
105
  {
106
+ "epoch": 0.42,
107
+ "grad_norm": 0.9688892960548401,
108
+ "learning_rate": 0.00017923322683706071,
109
+ "loss": 0.2867,
110
  "step": 130
111
  },
112
  {
113
+ "epoch": 0.45,
114
+ "grad_norm": 0.5258199572563171,
115
+ "learning_rate": 0.00017763578274760384,
116
+ "loss": 0.2557,
117
  "step": 140
118
  },
119
  {
120
+ "epoch": 0.48,
121
+ "grad_norm": 0.8469595313072205,
122
+ "learning_rate": 0.000176038338658147,
123
+ "loss": 0.3041,
124
  "step": 150
125
  },
126
  {
127
+ "epoch": 0.51,
128
+ "grad_norm": 1.846753716468811,
129
+ "learning_rate": 0.0001744408945686901,
130
+ "loss": 0.2766,
131
  "step": 160
132
  },
133
  {
134
+ "epoch": 0.54,
135
+ "grad_norm": 0.4423494338989258,
136
+ "learning_rate": 0.00017284345047923323,
137
+ "loss": 0.2555,
138
  "step": 170
139
  },
140
  {
141
+ "epoch": 0.58,
142
+ "grad_norm": 2.977750062942505,
143
+ "learning_rate": 0.00017124600638977638,
144
+ "loss": 0.2742,
145
  "step": 180
146
  },
147
  {
148
+ "epoch": 0.61,
149
+ "grad_norm": 0.8587075471878052,
150
+ "learning_rate": 0.00016964856230031948,
151
+ "loss": 0.3036,
152
  "step": 190
153
  },
154
  {
155
+ "epoch": 0.64,
156
+ "grad_norm": 0.7394629120826721,
157
+ "learning_rate": 0.00016805111821086263,
158
+ "loss": 0.2877,
159
  "step": 200
160
  },
161
  {
162
+ "epoch": 0.64,
163
+ "eval_loss": 0.2782871127128601,
164
+ "eval_na_accuracy": 0.944,
165
+ "eval_ordinal_accuracy": 0.5920745920745921,
166
+ "eval_runtime": 19.4606,
167
+ "eval_samples_per_second": 25.693,
168
+ "eval_steps_per_second": 3.237,
169
  "step": 200
170
  },
171
  {
172
+ "epoch": 0.67,
173
+ "grad_norm": 1.093324899673462,
174
+ "learning_rate": 0.00016645367412140575,
175
+ "loss": 0.2535,
176
  "step": 210
177
  },
178
  {
179
+ "epoch": 0.7,
180
+ "grad_norm": 1.0213277339935303,
181
+ "learning_rate": 0.0001648562300319489,
182
+ "loss": 0.2513,
183
  "step": 220
184
  },
185
  {
186
+ "epoch": 0.73,
187
+ "grad_norm": 1.1321722269058228,
188
+ "learning_rate": 0.00016325878594249202,
189
+ "loss": 0.3015,
190
  "step": 230
191
  },
192
  {
193
+ "epoch": 0.77,
194
+ "grad_norm": 3.1191928386688232,
195
+ "learning_rate": 0.00016166134185303515,
196
+ "loss": 0.2307,
197
  "step": 240
198
  },
199
  {
200
+ "epoch": 0.8,
201
+ "grad_norm": 1.107519507408142,
202
+ "learning_rate": 0.0001600638977635783,
203
+ "loss": 0.3198,
204
  "step": 250
205
  },
206
+ {
207
+ "epoch": 0.83,
208
+ "grad_norm": 0.818793535232544,
209
+ "learning_rate": 0.00015846645367412142,
210
+ "loss": 0.2959,
211
+ "step": 260
212
+ },
213
+ {
214
+ "epoch": 0.86,
215
+ "grad_norm": 0.4491446912288666,
216
+ "learning_rate": 0.00015686900958466454,
217
+ "loss": 0.212,
218
+ "step": 270
219
+ },
220
+ {
221
+ "epoch": 0.89,
222
+ "grad_norm": 1.838297963142395,
223
+ "learning_rate": 0.00015527156549520767,
224
+ "loss": 0.2836,
225
+ "step": 280
226
+ },
227
+ {
228
+ "epoch": 0.93,
229
+ "grad_norm": 1.4379059076309204,
230
+ "learning_rate": 0.00015367412140575082,
231
+ "loss": 0.309,
232
+ "step": 290
233
+ },
234
+ {
235
+ "epoch": 0.96,
236
+ "grad_norm": 2.274575710296631,
237
+ "learning_rate": 0.00015207667731629394,
238
+ "loss": 0.3079,
239
+ "step": 300
240
+ },
241
+ {
242
+ "epoch": 0.96,
243
+ "eval_loss": 0.2544197142124176,
244
+ "eval_na_accuracy": 0.958,
245
+ "eval_ordinal_accuracy": 0.5337995337995338,
246
+ "eval_runtime": 19.3178,
247
+ "eval_samples_per_second": 25.883,
248
+ "eval_steps_per_second": 3.261,
249
+ "step": 300
250
+ },
251
+ {
252
+ "epoch": 0.99,
253
+ "grad_norm": 0.7160611748695374,
254
+ "learning_rate": 0.00015047923322683706,
255
+ "loss": 0.223,
256
+ "step": 310
257
+ },
258
+ {
259
+ "epoch": 1.02,
260
+ "grad_norm": 0.5523737668991089,
261
+ "learning_rate": 0.0001488817891373802,
262
+ "loss": 0.2784,
263
+ "step": 320
264
+ },
265
+ {
266
+ "epoch": 1.05,
267
+ "grad_norm": 0.5268480777740479,
268
+ "learning_rate": 0.00014728434504792333,
269
+ "loss": 0.1671,
270
+ "step": 330
271
+ },
272
+ {
273
+ "epoch": 1.09,
274
+ "grad_norm": 1.7597477436065674,
275
+ "learning_rate": 0.00014568690095846646,
276
+ "loss": 0.1519,
277
+ "step": 340
278
+ },
279
+ {
280
+ "epoch": 1.12,
281
+ "grad_norm": 0.7296491861343384,
282
+ "learning_rate": 0.00014408945686900958,
283
+ "loss": 0.223,
284
+ "step": 350
285
+ },
286
+ {
287
+ "epoch": 1.15,
288
+ "grad_norm": 0.3579563796520233,
289
+ "learning_rate": 0.00014249201277955273,
290
+ "loss": 0.1383,
291
+ "step": 360
292
+ },
293
+ {
294
+ "epoch": 1.18,
295
+ "grad_norm": 1.5826038122177124,
296
+ "learning_rate": 0.00014089456869009585,
297
+ "loss": 0.2555,
298
+ "step": 370
299
+ },
300
+ {
301
+ "epoch": 1.21,
302
+ "grad_norm": 1.290208339691162,
303
+ "learning_rate": 0.000139297124600639,
304
+ "loss": 0.1482,
305
+ "step": 380
306
+ },
307
+ {
308
+ "epoch": 1.25,
309
+ "grad_norm": 0.9879806637763977,
310
+ "learning_rate": 0.00013769968051118212,
311
+ "loss": 0.1714,
312
+ "step": 390
313
+ },
314
+ {
315
+ "epoch": 1.28,
316
+ "grad_norm": 0.6665166616439819,
317
+ "learning_rate": 0.00013610223642172525,
318
+ "loss": 0.1901,
319
+ "step": 400
320
+ },
321
+ {
322
+ "epoch": 1.28,
323
+ "eval_loss": 0.26315367221832275,
324
+ "eval_na_accuracy": 0.948,
325
+ "eval_ordinal_accuracy": 0.6060606060606061,
326
+ "eval_runtime": 19.5095,
327
+ "eval_samples_per_second": 25.628,
328
+ "eval_steps_per_second": 3.229,
329
+ "step": 400
330
+ },
331
+ {
332
+ "epoch": 1.31,
333
+ "grad_norm": 0.5691338181495667,
334
+ "learning_rate": 0.00013450479233226837,
335
+ "loss": 0.21,
336
+ "step": 410
337
+ },
338
+ {
339
+ "epoch": 1.34,
340
+ "grad_norm": 0.76593017578125,
341
+ "learning_rate": 0.0001329073482428115,
342
+ "loss": 0.1704,
343
+ "step": 420
344
+ },
345
+ {
346
+ "epoch": 1.37,
347
+ "grad_norm": 0.6972767114639282,
348
+ "learning_rate": 0.00013130990415335464,
349
+ "loss": 0.1911,
350
+ "step": 430
351
+ },
352
+ {
353
+ "epoch": 1.41,
354
+ "grad_norm": 0.6530088186264038,
355
+ "learning_rate": 0.00012971246006389777,
356
+ "loss": 0.128,
357
+ "step": 440
358
+ },
359
+ {
360
+ "epoch": 1.44,
361
+ "grad_norm": 0.8316710591316223,
362
+ "learning_rate": 0.00012811501597444092,
363
+ "loss": 0.2046,
364
+ "step": 450
365
+ },
366
+ {
367
+ "epoch": 1.47,
368
+ "grad_norm": 0.5624408721923828,
369
+ "learning_rate": 0.00012651757188498404,
370
+ "loss": 0.1888,
371
+ "step": 460
372
+ },
373
+ {
374
+ "epoch": 1.5,
375
+ "grad_norm": 0.5218725800514221,
376
+ "learning_rate": 0.00012492012779552716,
377
+ "loss": 0.2096,
378
+ "step": 470
379
+ },
380
+ {
381
+ "epoch": 1.53,
382
+ "grad_norm": 0.7621105909347534,
383
+ "learning_rate": 0.00012332268370607028,
384
+ "loss": 0.1682,
385
+ "step": 480
386
+ },
387
+ {
388
+ "epoch": 1.57,
389
+ "grad_norm": 0.9870132803916931,
390
+ "learning_rate": 0.00012172523961661342,
391
+ "loss": 0.1606,
392
+ "step": 490
393
+ },
394
+ {
395
+ "epoch": 1.6,
396
+ "grad_norm": 0.5003547072410583,
397
+ "learning_rate": 0.00012012779552715656,
398
+ "loss": 0.277,
399
+ "step": 500
400
+ },
401
+ {
402
+ "epoch": 1.6,
403
+ "eval_loss": 0.24043463170528412,
404
+ "eval_na_accuracy": 0.96,
405
+ "eval_ordinal_accuracy": 0.5920745920745921,
406
+ "eval_runtime": 19.0104,
407
+ "eval_samples_per_second": 26.301,
408
+ "eval_steps_per_second": 3.314,
409
+ "step": 500
410
+ },
411
+ {
412
+ "epoch": 1.63,
413
+ "grad_norm": 2.554410457611084,
414
+ "learning_rate": 0.00011853035143769968,
415
+ "loss": 0.2163,
416
+ "step": 510
417
+ },
418
+ {
419
+ "epoch": 1.66,
420
+ "grad_norm": 0.577261745929718,
421
+ "learning_rate": 0.00011693290734824283,
422
+ "loss": 0.1652,
423
+ "step": 520
424
+ },
425
+ {
426
+ "epoch": 1.69,
427
+ "grad_norm": 0.722175121307373,
428
+ "learning_rate": 0.00011533546325878595,
429
+ "loss": 0.1061,
430
+ "step": 530
431
+ },
432
+ {
433
+ "epoch": 1.73,
434
+ "grad_norm": 2.013876438140869,
435
+ "learning_rate": 0.00011373801916932908,
436
+ "loss": 0.2078,
437
+ "step": 540
438
+ },
439
+ {
440
+ "epoch": 1.76,
441
+ "grad_norm": 2.5454280376434326,
442
+ "learning_rate": 0.00011214057507987221,
443
+ "loss": 0.238,
444
+ "step": 550
445
+ },
446
+ {
447
+ "epoch": 1.79,
448
+ "grad_norm": 0.7590613961219788,
449
+ "learning_rate": 0.00011054313099041533,
450
+ "loss": 0.1868,
451
+ "step": 560
452
+ },
453
+ {
454
+ "epoch": 1.82,
455
+ "grad_norm": 1.2678501605987549,
456
+ "learning_rate": 0.00010894568690095847,
457
+ "loss": 0.2365,
458
+ "step": 570
459
+ },
460
+ {
461
+ "epoch": 1.85,
462
+ "grad_norm": 0.9851756691932678,
463
+ "learning_rate": 0.0001073482428115016,
464
+ "loss": 0.2362,
465
+ "step": 580
466
+ },
467
+ {
468
+ "epoch": 1.88,
469
+ "grad_norm": 1.6259474754333496,
470
+ "learning_rate": 0.00010575079872204474,
471
+ "loss": 0.2368,
472
+ "step": 590
473
+ },
474
+ {
475
+ "epoch": 1.92,
476
+ "grad_norm": 0.7762842178344727,
477
+ "learning_rate": 0.00010415335463258787,
478
+ "loss": 0.2081,
479
+ "step": 600
480
+ },
481
+ {
482
+ "epoch": 1.92,
483
+ "eval_loss": 0.2585134506225586,
484
+ "eval_na_accuracy": 0.95,
485
+ "eval_ordinal_accuracy": 0.627039627039627,
486
+ "eval_runtime": 19.183,
487
+ "eval_samples_per_second": 26.065,
488
+ "eval_steps_per_second": 3.284,
489
+ "step": 600
490
+ },
491
+ {
492
+ "epoch": 1.95,
493
+ "grad_norm": 0.8312740325927734,
494
+ "learning_rate": 0.000102555910543131,
495
+ "loss": 0.1747,
496
+ "step": 610
497
+ },
498
+ {
499
+ "epoch": 1.98,
500
+ "grad_norm": 1.1986323595046997,
501
+ "learning_rate": 0.00010095846645367413,
502
+ "loss": 0.146,
503
+ "step": 620
504
+ },
505
+ {
506
+ "epoch": 2.01,
507
+ "grad_norm": 0.6852553486824036,
508
+ "learning_rate": 9.936102236421726e-05,
509
+ "loss": 0.1098,
510
+ "step": 630
511
+ },
512
+ {
513
+ "epoch": 2.04,
514
+ "grad_norm": 0.49778783321380615,
515
+ "learning_rate": 9.77635782747604e-05,
516
+ "loss": 0.0807,
517
+ "step": 640
518
+ },
519
+ {
520
+ "epoch": 2.08,
521
+ "grad_norm": 0.43836385011672974,
522
+ "learning_rate": 9.616613418530351e-05,
523
+ "loss": 0.0785,
524
+ "step": 650
525
+ },
526
+ {
527
+ "epoch": 2.11,
528
+ "grad_norm": 0.4496062695980072,
529
+ "learning_rate": 9.456869009584664e-05,
530
+ "loss": 0.1007,
531
+ "step": 660
532
+ },
533
+ {
534
+ "epoch": 2.14,
535
+ "grad_norm": 0.6552535891532898,
536
+ "learning_rate": 9.297124600638978e-05,
537
+ "loss": 0.0837,
538
+ "step": 670
539
+ },
540
+ {
541
+ "epoch": 2.17,
542
+ "grad_norm": 0.8565073013305664,
543
+ "learning_rate": 9.137380191693292e-05,
544
+ "loss": 0.1288,
545
+ "step": 680
546
+ },
547
+ {
548
+ "epoch": 2.2,
549
+ "grad_norm": 0.3386971056461334,
550
+ "learning_rate": 8.977635782747604e-05,
551
+ "loss": 0.0671,
552
+ "step": 690
553
+ },
554
+ {
555
+ "epoch": 2.24,
556
+ "grad_norm": 0.6647264957427979,
557
+ "learning_rate": 8.817891373801918e-05,
558
+ "loss": 0.0809,
559
+ "step": 700
560
+ },
561
+ {
562
+ "epoch": 2.24,
563
+ "eval_loss": 0.26309889554977417,
564
+ "eval_na_accuracy": 0.954,
565
+ "eval_ordinal_accuracy": 0.6736596736596736,
566
+ "eval_runtime": 18.6607,
567
+ "eval_samples_per_second": 26.794,
568
+ "eval_steps_per_second": 3.376,
569
+ "step": 700
570
+ },
571
+ {
572
+ "epoch": 2.27,
573
+ "grad_norm": 0.6069409251213074,
574
+ "learning_rate": 8.658146964856231e-05,
575
+ "loss": 0.0765,
576
+ "step": 710
577
+ },
578
+ {
579
+ "epoch": 2.3,
580
+ "grad_norm": 1.3801295757293701,
581
+ "learning_rate": 8.498402555910544e-05,
582
+ "loss": 0.1077,
583
+ "step": 720
584
+ },
585
+ {
586
+ "epoch": 2.33,
587
+ "grad_norm": 0.7153878808021545,
588
+ "learning_rate": 8.338658146964856e-05,
589
+ "loss": 0.075,
590
+ "step": 730
591
+ },
592
+ {
593
+ "epoch": 2.36,
594
+ "grad_norm": 0.727155327796936,
595
+ "learning_rate": 8.17891373801917e-05,
596
+ "loss": 0.0639,
597
+ "step": 740
598
+ },
599
+ {
600
+ "epoch": 2.4,
601
+ "grad_norm": 0.7251117825508118,
602
+ "learning_rate": 8.019169329073483e-05,
603
+ "loss": 0.1146,
604
+ "step": 750
605
+ },
606
+ {
607
+ "epoch": 2.43,
608
+ "grad_norm": 0.7786515951156616,
609
+ "learning_rate": 7.859424920127795e-05,
610
+ "loss": 0.1122,
611
+ "step": 760
612
+ },
613
+ {
614
+ "epoch": 2.46,
615
+ "grad_norm": 0.6191527843475342,
616
+ "learning_rate": 7.699680511182109e-05,
617
+ "loss": 0.0867,
618
+ "step": 770
619
+ },
620
+ {
621
+ "epoch": 2.49,
622
+ "grad_norm": 0.4263085722923279,
623
+ "learning_rate": 7.539936102236423e-05,
624
+ "loss": 0.1325,
625
+ "step": 780
626
+ },
627
+ {
628
+ "epoch": 2.52,
629
+ "grad_norm": 1.6009018421173096,
630
+ "learning_rate": 7.380191693290735e-05,
631
+ "loss": 0.0693,
632
+ "step": 790
633
+ },
634
+ {
635
+ "epoch": 2.56,
636
+ "grad_norm": 0.37769052386283875,
637
+ "learning_rate": 7.220447284345049e-05,
638
+ "loss": 0.1,
639
+ "step": 800
640
+ },
641
+ {
642
+ "epoch": 2.56,
643
+ "eval_loss": 0.2692907154560089,
644
+ "eval_na_accuracy": 0.958,
645
+ "eval_ordinal_accuracy": 0.6363636363636364,
646
+ "eval_runtime": 19.3788,
647
+ "eval_samples_per_second": 25.801,
648
+ "eval_steps_per_second": 3.251,
649
+ "step": 800
650
+ },
651
+ {
652
+ "epoch": 2.59,
653
+ "grad_norm": 0.45006510615348816,
654
+ "learning_rate": 7.060702875399361e-05,
655
+ "loss": 0.0645,
656
+ "step": 810
657
+ },
658
+ {
659
+ "epoch": 2.62,
660
+ "grad_norm": 0.813955545425415,
661
+ "learning_rate": 6.900958466453674e-05,
662
+ "loss": 0.0583,
663
+ "step": 820
664
+ },
665
+ {
666
+ "epoch": 2.65,
667
+ "grad_norm": 0.7967355847358704,
668
+ "learning_rate": 6.741214057507987e-05,
669
+ "loss": 0.1065,
670
+ "step": 830
671
+ },
672
+ {
673
+ "epoch": 2.68,
674
+ "grad_norm": 0.8020057678222656,
675
+ "learning_rate": 6.5814696485623e-05,
676
+ "loss": 0.0821,
677
+ "step": 840
678
+ },
679
+ {
680
+ "epoch": 2.72,
681
+ "grad_norm": 0.7547793388366699,
682
+ "learning_rate": 6.421725239616614e-05,
683
+ "loss": 0.0983,
684
+ "step": 850
685
+ },
686
+ {
687
+ "epoch": 2.75,
688
+ "grad_norm": 0.42487284541130066,
689
+ "learning_rate": 6.261980830670928e-05,
690
+ "loss": 0.1221,
691
+ "step": 860
692
+ },
693
+ {
694
+ "epoch": 2.78,
695
+ "grad_norm": 0.3974541425704956,
696
+ "learning_rate": 6.1022364217252406e-05,
697
+ "loss": 0.108,
698
+ "step": 870
699
+ },
700
+ {
701
+ "epoch": 2.81,
702
+ "grad_norm": 0.42565950751304626,
703
+ "learning_rate": 5.942492012779552e-05,
704
+ "loss": 0.106,
705
+ "step": 880
706
+ },
707
+ {
708
+ "epoch": 2.84,
709
+ "grad_norm": 0.5634491443634033,
710
+ "learning_rate": 5.782747603833866e-05,
711
+ "loss": 0.0765,
712
+ "step": 890
713
+ },
714
+ {
715
+ "epoch": 2.88,
716
+ "grad_norm": 0.7806987166404724,
717
+ "learning_rate": 5.623003194888179e-05,
718
+ "loss": 0.0924,
719
+ "step": 900
720
+ },
721
+ {
722
+ "epoch": 2.88,
723
+ "eval_loss": 0.2566128671169281,
724
+ "eval_na_accuracy": 0.962,
725
+ "eval_ordinal_accuracy": 0.6503496503496503,
726
+ "eval_runtime": 19.1674,
727
+ "eval_samples_per_second": 26.086,
728
+ "eval_steps_per_second": 3.287,
729
+ "step": 900
730
+ },
731
+ {
732
+ "epoch": 2.91,
733
+ "grad_norm": 1.5311517715454102,
734
+ "learning_rate": 5.4632587859424925e-05,
735
+ "loss": 0.0779,
736
+ "step": 910
737
+ },
738
+ {
739
+ "epoch": 2.94,
740
+ "grad_norm": 0.8630849123001099,
741
+ "learning_rate": 5.3035143769968054e-05,
742
+ "loss": 0.0697,
743
+ "step": 920
744
+ },
745
+ {
746
+ "epoch": 2.97,
747
+ "grad_norm": 0.39383065700531006,
748
+ "learning_rate": 5.1437699680511184e-05,
749
+ "loss": 0.065,
750
+ "step": 930
751
+ },
752
+ {
753
+ "epoch": 3.0,
754
+ "grad_norm": 0.45358964800834656,
755
+ "learning_rate": 4.984025559105431e-05,
756
+ "loss": 0.0489,
757
+ "step": 940
758
+ },
759
+ {
760
+ "epoch": 3.04,
761
+ "grad_norm": 0.39319777488708496,
762
+ "learning_rate": 4.824281150159744e-05,
763
+ "loss": 0.0311,
764
+ "step": 950
765
+ },
766
+ {
767
+ "epoch": 3.07,
768
+ "grad_norm": 0.6434493064880371,
769
+ "learning_rate": 4.664536741214058e-05,
770
+ "loss": 0.0424,
771
+ "step": 960
772
+ },
773
+ {
774
+ "epoch": 3.1,
775
+ "grad_norm": 0.4773056209087372,
776
+ "learning_rate": 4.504792332268371e-05,
777
+ "loss": 0.0326,
778
+ "step": 970
779
+ },
780
+ {
781
+ "epoch": 3.13,
782
+ "grad_norm": 0.48347094655036926,
783
+ "learning_rate": 4.345047923322684e-05,
784
+ "loss": 0.0312,
785
+ "step": 980
786
+ },
787
+ {
788
+ "epoch": 3.16,
789
+ "grad_norm": 0.410236120223999,
790
+ "learning_rate": 4.185303514376997e-05,
791
+ "loss": 0.0283,
792
+ "step": 990
793
+ },
794
+ {
795
+ "epoch": 3.19,
796
+ "grad_norm": 0.3112216889858246,
797
+ "learning_rate": 4.0255591054313104e-05,
798
+ "loss": 0.0304,
799
+ "step": 1000
800
+ },
801
+ {
802
+ "epoch": 3.19,
803
+ "eval_loss": 0.2782021760940552,
804
+ "eval_na_accuracy": 0.954,
805
+ "eval_ordinal_accuracy": 0.6526806526806527,
806
+ "eval_runtime": 18.9723,
807
+ "eval_samples_per_second": 26.354,
808
+ "eval_steps_per_second": 3.321,
809
+ "step": 1000
810
+ },
811
+ {
812
+ "epoch": 3.23,
813
+ "grad_norm": 0.3799457550048828,
814
+ "learning_rate": 3.8658146964856234e-05,
815
+ "loss": 0.0622,
816
+ "step": 1010
817
+ },
818
+ {
819
+ "epoch": 3.26,
820
+ "grad_norm": 0.3940556049346924,
821
+ "learning_rate": 3.7060702875399364e-05,
822
+ "loss": 0.0275,
823
+ "step": 1020
824
+ },
825
+ {
826
+ "epoch": 3.29,
827
+ "grad_norm": 0.5390946269035339,
828
+ "learning_rate": 3.546325878594249e-05,
829
+ "loss": 0.0531,
830
+ "step": 1030
831
+ },
832
+ {
833
+ "epoch": 3.32,
834
+ "grad_norm": 0.5048585534095764,
835
+ "learning_rate": 3.386581469648562e-05,
836
+ "loss": 0.03,
837
+ "step": 1040
838
+ },
839
+ {
840
+ "epoch": 3.35,
841
+ "grad_norm": 0.4280671775341034,
842
+ "learning_rate": 3.242811501597444e-05,
843
+ "loss": 0.0474,
844
+ "step": 1050
845
+ },
846
+ {
847
+ "epoch": 3.39,
848
+ "grad_norm": 0.34902918338775635,
849
+ "learning_rate": 3.083067092651757e-05,
850
+ "loss": 0.0318,
851
+ "step": 1060
852
+ },
853
+ {
854
+ "epoch": 3.42,
855
+ "grad_norm": 0.4760427176952362,
856
+ "learning_rate": 2.9233226837060707e-05,
857
+ "loss": 0.0285,
858
+ "step": 1070
859
+ },
860
+ {
861
+ "epoch": 3.45,
862
+ "grad_norm": 0.15729285776615143,
863
+ "learning_rate": 2.7635782747603834e-05,
864
+ "loss": 0.0496,
865
+ "step": 1080
866
+ },
867
+ {
868
+ "epoch": 3.48,
869
+ "grad_norm": 0.38401302695274353,
870
+ "learning_rate": 2.6038338658146967e-05,
871
+ "loss": 0.0252,
872
+ "step": 1090
873
+ },
874
+ {
875
+ "epoch": 3.51,
876
+ "grad_norm": 0.3598209023475647,
877
+ "learning_rate": 2.44408945686901e-05,
878
+ "loss": 0.0256,
879
+ "step": 1100
880
+ },
881
+ {
882
+ "epoch": 3.51,
883
+ "eval_loss": 0.26621493697166443,
884
+ "eval_na_accuracy": 0.96,
885
+ "eval_ordinal_accuracy": 0.6643356643356644,
886
+ "eval_runtime": 19.2883,
887
+ "eval_samples_per_second": 25.922,
888
+ "eval_steps_per_second": 3.266,
889
+ "step": 1100
890
+ },
891
+ {
892
+ "epoch": 3.55,
893
+ "grad_norm": 0.4313770830631256,
894
+ "learning_rate": 2.284345047923323e-05,
895
+ "loss": 0.0269,
896
+ "step": 1110
897
+ },
898
+ {
899
+ "epoch": 3.58,
900
+ "grad_norm": 0.4230475425720215,
901
+ "learning_rate": 2.124600638977636e-05,
902
+ "loss": 0.0304,
903
+ "step": 1120
904
+ },
905
+ {
906
+ "epoch": 3.61,
907
+ "grad_norm": 0.2662275433540344,
908
+ "learning_rate": 1.964856230031949e-05,
909
+ "loss": 0.0176,
910
+ "step": 1130
911
+ },
912
+ {
913
+ "epoch": 3.64,
914
+ "grad_norm": 0.791589617729187,
915
+ "learning_rate": 1.805111821086262e-05,
916
+ "loss": 0.0307,
917
+ "step": 1140
918
+ },
919
+ {
920
+ "epoch": 3.67,
921
+ "grad_norm": 0.49746203422546387,
922
+ "learning_rate": 1.645367412140575e-05,
923
+ "loss": 0.0217,
924
+ "step": 1150
925
+ },
926
+ {
927
+ "epoch": 3.71,
928
+ "grad_norm": 0.18784604966640472,
929
+ "learning_rate": 1.485623003194888e-05,
930
+ "loss": 0.0206,
931
+ "step": 1160
932
+ },
933
+ {
934
+ "epoch": 3.74,
935
+ "grad_norm": 0.47459328174591064,
936
+ "learning_rate": 1.3258785942492014e-05,
937
+ "loss": 0.0265,
938
+ "step": 1170
939
+ },
940
+ {
941
+ "epoch": 3.77,
942
+ "grad_norm": 0.2313966453075409,
943
+ "learning_rate": 1.1661341853035145e-05,
944
+ "loss": 0.0313,
945
+ "step": 1180
946
+ },
947
+ {
948
+ "epoch": 3.8,
949
+ "grad_norm": 0.4523174464702606,
950
+ "learning_rate": 1.0063897763578276e-05,
951
+ "loss": 0.0622,
952
+ "step": 1190
953
+ },
954
+ {
955
+ "epoch": 3.83,
956
+ "grad_norm": 0.17250552773475647,
957
+ "learning_rate": 8.466453674121406e-06,
958
+ "loss": 0.0424,
959
+ "step": 1200
960
+ },
961
+ {
962
+ "epoch": 3.83,
963
+ "eval_loss": 0.2670270502567291,
964
+ "eval_na_accuracy": 0.962,
965
+ "eval_ordinal_accuracy": 0.675990675990676,
966
+ "eval_runtime": 18.6815,
967
+ "eval_samples_per_second": 26.764,
968
+ "eval_steps_per_second": 3.372,
969
+ "step": 1200
970
+ },
971
+ {
972
+ "epoch": 3.87,
973
+ "grad_norm": 0.3159545660018921,
974
+ "learning_rate": 6.869009584664538e-06,
975
+ "loss": 0.0209,
976
+ "step": 1210
977
+ },
978
+ {
979
+ "epoch": 3.9,
980
+ "grad_norm": 0.5657308101654053,
981
+ "learning_rate": 5.2715654952076674e-06,
982
+ "loss": 0.019,
983
+ "step": 1220
984
+ },
985
+ {
986
+ "epoch": 3.93,
987
+ "grad_norm": 0.4505397379398346,
988
+ "learning_rate": 3.6741214057507987e-06,
989
+ "loss": 0.0249,
990
+ "step": 1230
991
+ },
992
+ {
993
+ "epoch": 3.96,
994
+ "grad_norm": 0.21818359196186066,
995
+ "learning_rate": 2.0766773162939296e-06,
996
+ "loss": 0.0226,
997
+ "step": 1240
998
+ },
999
+ {
1000
+ "epoch": 3.99,
1001
+ "grad_norm": 0.28112998604774475,
1002
+ "learning_rate": 4.792332268370607e-07,
1003
+ "loss": 0.027,
1004
+ "step": 1250
1005
+ },
1006
  {
1007
  "epoch": 4.0,
1008
+ "step": 1252,
1009
+ "total_flos": 1.5498953551872e+18,
1010
+ "train_loss": 0.15608444792060808,
1011
+ "train_runtime": 1745.0942,
1012
+ "train_samples_per_second": 11.461,
1013
+ "train_steps_per_second": 0.717
1014
  }
1015
  ],
1016
  "logging_steps": 10,
1017
+ "max_steps": 1252,
1018
  "num_input_tokens_seen": 0,
1019
  "num_train_epochs": 4,
1020
  "save_steps": 100,
1021
+ "total_flos": 1.5498953551872e+18,
1022
  "train_batch_size": 16,
1023
  "trial_name": null,
1024
  "trial_params": null