End of training
Browse files- README.md +2 -2
- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +526 -0
README.md
CHANGED
|
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [MCG-NJU/videomae-base-finetuned-kinetics](https://huggingface.co/MCG-NJU/videomae-base-finetuned-kinetics) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 1.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [MCG-NJU/videomae-base-finetuned-kinetics](https://huggingface.co/MCG-NJU/videomae-base-finetuned-kinetics) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.2269
|
| 21 |
+
- Accuracy: 0.5443
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.02,
|
| 3 |
+
"eval_accuracy": 0.5443037974683544,
|
| 4 |
+
"eval_loss": 1.2269208431243896,
|
| 5 |
+
"eval_runtime": 27.3703,
|
| 6 |
+
"eval_samples_per_second": 2.886,
|
| 7 |
+
"eval_steps_per_second": 0.512
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.02,
|
| 3 |
+
"eval_accuracy": 0.5443037974683544,
|
| 4 |
+
"eval_loss": 1.2269208431243896,
|
| 5 |
+
"eval_runtime": 27.3703,
|
| 6 |
+
"eval_samples_per_second": 2.886,
|
| 7 |
+
"eval_steps_per_second": 0.512
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,526 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5443037974683544,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-caer-final-4-epochs\\checkpoint-246",
|
| 4 |
+
"epoch": 7.017123287671233,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 584,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 6.458174705505371,
|
| 14 |
+
"learning_rate": 8.47457627118644e-06,
|
| 15 |
+
"loss": 1.602,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03,
|
| 20 |
+
"grad_norm": 7.2191619873046875,
|
| 21 |
+
"learning_rate": 1.694915254237288e-05,
|
| 22 |
+
"loss": 1.6082,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05,
|
| 27 |
+
"grad_norm": 6.413382053375244,
|
| 28 |
+
"learning_rate": 2.5423728813559322e-05,
|
| 29 |
+
"loss": 1.6091,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07,
|
| 34 |
+
"grad_norm": 6.1725263595581055,
|
| 35 |
+
"learning_rate": 3.389830508474576e-05,
|
| 36 |
+
"loss": 1.5506,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09,
|
| 41 |
+
"grad_norm": 6.784051418304443,
|
| 42 |
+
"learning_rate": 4.2372881355932206e-05,
|
| 43 |
+
"loss": 1.5818,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1,
|
| 48 |
+
"grad_norm": 6.475350856781006,
|
| 49 |
+
"learning_rate": 4.990476190476191e-05,
|
| 50 |
+
"loss": 1.5237,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.12,
|
| 55 |
+
"grad_norm": 8.449509620666504,
|
| 56 |
+
"learning_rate": 4.895238095238096e-05,
|
| 57 |
+
"loss": 1.6543,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14,
|
| 62 |
+
"grad_norm": 5.603182792663574,
|
| 63 |
+
"learning_rate": 4.8e-05,
|
| 64 |
+
"loss": 1.3388,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.14,
|
| 69 |
+
"eval_accuracy": 0.3924050632911392,
|
| 70 |
+
"eval_loss": 1.4801665544509888,
|
| 71 |
+
"eval_runtime": 25.4554,
|
| 72 |
+
"eval_samples_per_second": 3.103,
|
| 73 |
+
"eval_steps_per_second": 0.55,
|
| 74 |
+
"step": 82
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.01,
|
| 78 |
+
"grad_norm": 8.551097869873047,
|
| 79 |
+
"learning_rate": 4.704761904761905e-05,
|
| 80 |
+
"loss": 1.4052,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.03,
|
| 85 |
+
"grad_norm": 11.188709259033203,
|
| 86 |
+
"learning_rate": 4.60952380952381e-05,
|
| 87 |
+
"loss": 1.345,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.05,
|
| 92 |
+
"grad_norm": 9.687131881713867,
|
| 93 |
+
"learning_rate": 4.514285714285714e-05,
|
| 94 |
+
"loss": 1.3573,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.07,
|
| 99 |
+
"grad_norm": 10.117310523986816,
|
| 100 |
+
"learning_rate": 4.419047619047619e-05,
|
| 101 |
+
"loss": 1.1564,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.08,
|
| 106 |
+
"grad_norm": 7.854645729064941,
|
| 107 |
+
"learning_rate": 4.323809523809524e-05,
|
| 108 |
+
"loss": 1.2247,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.1,
|
| 113 |
+
"grad_norm": 7.707760334014893,
|
| 114 |
+
"learning_rate": 4.228571428571429e-05,
|
| 115 |
+
"loss": 1.1877,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.12,
|
| 120 |
+
"grad_norm": 8.141950607299805,
|
| 121 |
+
"learning_rate": 4.133333333333333e-05,
|
| 122 |
+
"loss": 1.0836,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.13,
|
| 127 |
+
"grad_norm": 9.751481056213379,
|
| 128 |
+
"learning_rate": 4.038095238095238e-05,
|
| 129 |
+
"loss": 1.1881,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.14,
|
| 134 |
+
"eval_accuracy": 0.3670886075949367,
|
| 135 |
+
"eval_loss": 1.3925659656524658,
|
| 136 |
+
"eval_runtime": 25.2241,
|
| 137 |
+
"eval_samples_per_second": 3.132,
|
| 138 |
+
"eval_steps_per_second": 0.555,
|
| 139 |
+
"step": 164
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.01,
|
| 143 |
+
"grad_norm": 13.817303657531738,
|
| 144 |
+
"learning_rate": 3.942857142857143e-05,
|
| 145 |
+
"loss": 0.8556,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.03,
|
| 150 |
+
"grad_norm": 12.339946746826172,
|
| 151 |
+
"learning_rate": 3.847619047619048e-05,
|
| 152 |
+
"loss": 0.862,
|
| 153 |
+
"step": 180
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.04,
|
| 157 |
+
"grad_norm": 14.686485290527344,
|
| 158 |
+
"learning_rate": 3.752380952380953e-05,
|
| 159 |
+
"loss": 0.8869,
|
| 160 |
+
"step": 190
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.06,
|
| 164 |
+
"grad_norm": 9.216782569885254,
|
| 165 |
+
"learning_rate": 3.6571428571428576e-05,
|
| 166 |
+
"loss": 0.9421,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.08,
|
| 171 |
+
"grad_norm": 12.398260116577148,
|
| 172 |
+
"learning_rate": 3.561904761904762e-05,
|
| 173 |
+
"loss": 0.695,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.1,
|
| 178 |
+
"grad_norm": 11.731080055236816,
|
| 179 |
+
"learning_rate": 3.466666666666667e-05,
|
| 180 |
+
"loss": 1.0937,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.11,
|
| 185 |
+
"grad_norm": 10.27041244506836,
|
| 186 |
+
"learning_rate": 3.3714285714285716e-05,
|
| 187 |
+
"loss": 0.9285,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.13,
|
| 192 |
+
"grad_norm": 15.88553524017334,
|
| 193 |
+
"learning_rate": 3.276190476190477e-05,
|
| 194 |
+
"loss": 0.9722,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.14,
|
| 199 |
+
"eval_accuracy": 0.5443037974683544,
|
| 200 |
+
"eval_loss": 1.2269208431243896,
|
| 201 |
+
"eval_runtime": 26.6893,
|
| 202 |
+
"eval_samples_per_second": 2.96,
|
| 203 |
+
"eval_steps_per_second": 0.525,
|
| 204 |
+
"step": 246
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.01,
|
| 208 |
+
"grad_norm": 10.362627029418945,
|
| 209 |
+
"learning_rate": 3.180952380952381e-05,
|
| 210 |
+
"loss": 0.9198,
|
| 211 |
+
"step": 250
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.02,
|
| 215 |
+
"grad_norm": 7.394880771636963,
|
| 216 |
+
"learning_rate": 3.0857142857142856e-05,
|
| 217 |
+
"loss": 0.5962,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.04,
|
| 222 |
+
"grad_norm": 7.7396345138549805,
|
| 223 |
+
"learning_rate": 2.9904761904761907e-05,
|
| 224 |
+
"loss": 0.5537,
|
| 225 |
+
"step": 270
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.06,
|
| 229 |
+
"grad_norm": 15.393024444580078,
|
| 230 |
+
"learning_rate": 2.8952380952380952e-05,
|
| 231 |
+
"loss": 0.6155,
|
| 232 |
+
"step": 280
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.08,
|
| 236 |
+
"grad_norm": 13.79691219329834,
|
| 237 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 238 |
+
"loss": 0.7206,
|
| 239 |
+
"step": 290
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.09,
|
| 243 |
+
"grad_norm": 11.387887954711914,
|
| 244 |
+
"learning_rate": 2.704761904761905e-05,
|
| 245 |
+
"loss": 0.5823,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.11,
|
| 250 |
+
"grad_norm": 19.518312454223633,
|
| 251 |
+
"learning_rate": 2.6095238095238095e-05,
|
| 252 |
+
"loss": 0.7304,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.13,
|
| 257 |
+
"grad_norm": 8.372328758239746,
|
| 258 |
+
"learning_rate": 2.5142857142857147e-05,
|
| 259 |
+
"loss": 0.6545,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.14,
|
| 264 |
+
"eval_accuracy": 0.4936708860759494,
|
| 265 |
+
"eval_loss": 1.229438066482544,
|
| 266 |
+
"eval_runtime": 27.627,
|
| 267 |
+
"eval_samples_per_second": 2.86,
|
| 268 |
+
"eval_steps_per_second": 0.507,
|
| 269 |
+
"step": 328
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 4.0,
|
| 273 |
+
"grad_norm": 9.884891510009766,
|
| 274 |
+
"learning_rate": 2.419047619047619e-05,
|
| 275 |
+
"loss": 0.6225,
|
| 276 |
+
"step": 330
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.02,
|
| 280 |
+
"grad_norm": 24.8238525390625,
|
| 281 |
+
"learning_rate": 2.323809523809524e-05,
|
| 282 |
+
"loss": 0.4376,
|
| 283 |
+
"step": 340
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 4.04,
|
| 287 |
+
"grad_norm": 11.315938949584961,
|
| 288 |
+
"learning_rate": 2.2285714285714287e-05,
|
| 289 |
+
"loss": 0.4116,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.05,
|
| 294 |
+
"grad_norm": 12.32546615600586,
|
| 295 |
+
"learning_rate": 2.1333333333333335e-05,
|
| 296 |
+
"loss": 0.4504,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.07,
|
| 301 |
+
"grad_norm": 7.664936065673828,
|
| 302 |
+
"learning_rate": 2.0380952380952382e-05,
|
| 303 |
+
"loss": 0.6834,
|
| 304 |
+
"step": 370
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.09,
|
| 308 |
+
"grad_norm": 16.222686767578125,
|
| 309 |
+
"learning_rate": 1.942857142857143e-05,
|
| 310 |
+
"loss": 0.5456,
|
| 311 |
+
"step": 380
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 4.11,
|
| 315 |
+
"grad_norm": 7.430470943450928,
|
| 316 |
+
"learning_rate": 1.8476190476190478e-05,
|
| 317 |
+
"loss": 0.4284,
|
| 318 |
+
"step": 390
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.12,
|
| 322 |
+
"grad_norm": 2.954035520553589,
|
| 323 |
+
"learning_rate": 1.7523809523809526e-05,
|
| 324 |
+
"loss": 0.4361,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 4.14,
|
| 329 |
+
"grad_norm": 27.79668426513672,
|
| 330 |
+
"learning_rate": 1.657142857142857e-05,
|
| 331 |
+
"loss": 0.5289,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 4.14,
|
| 336 |
+
"eval_accuracy": 0.4810126582278481,
|
| 337 |
+
"eval_loss": 1.3317660093307495,
|
| 338 |
+
"eval_runtime": 27.2349,
|
| 339 |
+
"eval_samples_per_second": 2.901,
|
| 340 |
+
"eval_steps_per_second": 0.514,
|
| 341 |
+
"step": 410
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.02,
|
| 345 |
+
"grad_norm": 6.09324836730957,
|
| 346 |
+
"learning_rate": 1.561904761904762e-05,
|
| 347 |
+
"loss": 0.3316,
|
| 348 |
+
"step": 420
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.03,
|
| 352 |
+
"grad_norm": 13.406198501586914,
|
| 353 |
+
"learning_rate": 1.4666666666666668e-05,
|
| 354 |
+
"loss": 0.2502,
|
| 355 |
+
"step": 430
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 5.05,
|
| 359 |
+
"grad_norm": 10.042999267578125,
|
| 360 |
+
"learning_rate": 1.3714285714285716e-05,
|
| 361 |
+
"loss": 0.4386,
|
| 362 |
+
"step": 440
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 5.07,
|
| 366 |
+
"grad_norm": 10.368085861206055,
|
| 367 |
+
"learning_rate": 1.2761904761904764e-05,
|
| 368 |
+
"loss": 0.3197,
|
| 369 |
+
"step": 450
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.09,
|
| 373 |
+
"grad_norm": 14.689448356628418,
|
| 374 |
+
"learning_rate": 1.180952380952381e-05,
|
| 375 |
+
"loss": 0.4345,
|
| 376 |
+
"step": 460
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 5.1,
|
| 380 |
+
"grad_norm": 5.982275009155273,
|
| 381 |
+
"learning_rate": 1.0857142857142858e-05,
|
| 382 |
+
"loss": 0.3263,
|
| 383 |
+
"step": 470
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 5.12,
|
| 387 |
+
"grad_norm": 12.9437255859375,
|
| 388 |
+
"learning_rate": 9.904761904761906e-06,
|
| 389 |
+
"loss": 0.3721,
|
| 390 |
+
"step": 480
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.14,
|
| 394 |
+
"grad_norm": 17.087923049926758,
|
| 395 |
+
"learning_rate": 8.952380952380952e-06,
|
| 396 |
+
"loss": 0.2826,
|
| 397 |
+
"step": 490
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 5.14,
|
| 401 |
+
"eval_accuracy": 0.4810126582278481,
|
| 402 |
+
"eval_loss": 1.254180908203125,
|
| 403 |
+
"eval_runtime": 27.4561,
|
| 404 |
+
"eval_samples_per_second": 2.877,
|
| 405 |
+
"eval_steps_per_second": 0.51,
|
| 406 |
+
"step": 492
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 6.01,
|
| 410 |
+
"grad_norm": 11.193989753723145,
|
| 411 |
+
"learning_rate": 8.000000000000001e-06,
|
| 412 |
+
"loss": 0.1963,
|
| 413 |
+
"step": 500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.03,
|
| 417 |
+
"grad_norm": 2.909717559814453,
|
| 418 |
+
"learning_rate": 7.0476190476190475e-06,
|
| 419 |
+
"loss": 0.1711,
|
| 420 |
+
"step": 510
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 6.05,
|
| 424 |
+
"grad_norm": 14.048554420471191,
|
| 425 |
+
"learning_rate": 6.095238095238095e-06,
|
| 426 |
+
"loss": 0.1742,
|
| 427 |
+
"step": 520
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 6.07,
|
| 431 |
+
"grad_norm": 14.992708206176758,
|
| 432 |
+
"learning_rate": 5.142857142857143e-06,
|
| 433 |
+
"loss": 0.3157,
|
| 434 |
+
"step": 530
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.08,
|
| 438 |
+
"grad_norm": 13.224514961242676,
|
| 439 |
+
"learning_rate": 4.19047619047619e-06,
|
| 440 |
+
"loss": 0.2316,
|
| 441 |
+
"step": 540
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.1,
|
| 445 |
+
"grad_norm": 20.14113998413086,
|
| 446 |
+
"learning_rate": 3.238095238095238e-06,
|
| 447 |
+
"loss": 0.1789,
|
| 448 |
+
"step": 550
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.12,
|
| 452 |
+
"grad_norm": 5.539999008178711,
|
| 453 |
+
"learning_rate": 2.285714285714286e-06,
|
| 454 |
+
"loss": 0.1952,
|
| 455 |
+
"step": 560
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.13,
|
| 459 |
+
"grad_norm": 1.642961025238037,
|
| 460 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 461 |
+
"loss": 0.2948,
|
| 462 |
+
"step": 570
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 6.14,
|
| 466 |
+
"eval_accuracy": 0.5443037974683544,
|
| 467 |
+
"eval_loss": 1.2192754745483398,
|
| 468 |
+
"eval_runtime": 27.124,
|
| 469 |
+
"eval_samples_per_second": 2.913,
|
| 470 |
+
"eval_steps_per_second": 0.516,
|
| 471 |
+
"step": 574
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.01,
|
| 475 |
+
"grad_norm": 1.9858059883117676,
|
| 476 |
+
"learning_rate": 3.8095238095238096e-07,
|
| 477 |
+
"loss": 0.1954,
|
| 478 |
+
"step": 580
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 7.02,
|
| 482 |
+
"eval_accuracy": 0.5316455696202531,
|
| 483 |
+
"eval_loss": 1.2165864706039429,
|
| 484 |
+
"eval_runtime": 26.4535,
|
| 485 |
+
"eval_samples_per_second": 2.986,
|
| 486 |
+
"eval_steps_per_second": 0.529,
|
| 487 |
+
"step": 584
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 7.02,
|
| 491 |
+
"step": 584,
|
| 492 |
+
"total_flos": 4.331436394978345e+18,
|
| 493 |
+
"train_loss": 0.7601832610695329,
|
| 494 |
+
"train_runtime": 1891.5972,
|
| 495 |
+
"train_samples_per_second": 1.852,
|
| 496 |
+
"train_steps_per_second": 0.309
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 7.02,
|
| 500 |
+
"eval_accuracy": 0.4642857142857143,
|
| 501 |
+
"eval_loss": 1.2583760023117065,
|
| 502 |
+
"eval_runtime": 18.8662,
|
| 503 |
+
"eval_samples_per_second": 2.968,
|
| 504 |
+
"eval_steps_per_second": 0.53,
|
| 505 |
+
"step": 584
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 7.02,
|
| 509 |
+
"eval_accuracy": 0.5443037974683544,
|
| 510 |
+
"eval_loss": 1.2269208431243896,
|
| 511 |
+
"eval_runtime": 27.3703,
|
| 512 |
+
"eval_samples_per_second": 2.886,
|
| 513 |
+
"eval_steps_per_second": 0.512,
|
| 514 |
+
"step": 584
|
| 515 |
+
}
|
| 516 |
+
],
|
| 517 |
+
"logging_steps": 10,
|
| 518 |
+
"max_steps": 584,
|
| 519 |
+
"num_input_tokens_seen": 0,
|
| 520 |
+
"num_train_epochs": 9223372036854775807,
|
| 521 |
+
"save_steps": 500,
|
| 522 |
+
"total_flos": 4.331436394978345e+18,
|
| 523 |
+
"train_batch_size": 6,
|
| 524 |
+
"trial_name": null,
|
| 525 |
+
"trial_params": null
|
| 526 |
+
}
|