End of training
Browse files- README.md +1 -1
- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +642 -0
README.md
CHANGED
|
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 1.
|
| 21 |
- Accuracy: 0.3481
|
| 22 |
|
| 23 |
## Model description
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.7031
|
| 21 |
- Accuracy: 0.3481
|
| 22 |
|
| 23 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.09,
|
| 3 |
+
"eval_accuracy": 0.34811715481171546,
|
| 4 |
+
"eval_loss": 1.7031110525131226,
|
| 5 |
+
"eval_runtime": 266.484,
|
| 6 |
+
"eval_samples_per_second": 4.484,
|
| 7 |
+
"eval_steps_per_second": 0.281
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.09,
|
| 3 |
+
"eval_accuracy": 0.34811715481171546,
|
| 4 |
+
"eval_loss": 1.7031110525131226,
|
| 5 |
+
"eval_runtime": 266.484,
|
| 6 |
+
"eval_samples_per_second": 4.484,
|
| 7 |
+
"eval_steps_per_second": 0.281
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,642 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.34081346423562414,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-elderf1/checkpoint-73",
|
| 4 |
+
"epoch": 9.0875,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 720,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 4.3531270027160645,
|
| 14 |
+
"learning_rate": 0.0001388888888888889,
|
| 15 |
+
"loss": 1.7775,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03,
|
| 20 |
+
"grad_norm": 5.537198543548584,
|
| 21 |
+
"learning_rate": 0.0002777777777777778,
|
| 22 |
+
"loss": 1.6877,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04,
|
| 27 |
+
"grad_norm": 5.151556968688965,
|
| 28 |
+
"learning_rate": 0.0004166666666666667,
|
| 29 |
+
"loss": 1.7997,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06,
|
| 34 |
+
"grad_norm": 5.214244365692139,
|
| 35 |
+
"learning_rate": 0.0005555555555555556,
|
| 36 |
+
"loss": 1.8051,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07,
|
| 41 |
+
"grad_norm": 1.6136474609375,
|
| 42 |
+
"learning_rate": 0.0006944444444444445,
|
| 43 |
+
"loss": 1.7065,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08,
|
| 48 |
+
"grad_norm": 5.423031330108643,
|
| 49 |
+
"learning_rate": 0.0008333333333333334,
|
| 50 |
+
"loss": 1.7922,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1,
|
| 55 |
+
"grad_norm": 3.5037877559661865,
|
| 56 |
+
"learning_rate": 0.0009722222222222222,
|
| 57 |
+
"loss": 1.7358,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1,
|
| 62 |
+
"eval_accuracy": 0.34081346423562414,
|
| 63 |
+
"eval_loss": 1.692335844039917,
|
| 64 |
+
"eval_runtime": 379.7283,
|
| 65 |
+
"eval_samples_per_second": 3.755,
|
| 66 |
+
"eval_steps_per_second": 0.237,
|
| 67 |
+
"step": 73
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.01,
|
| 71 |
+
"grad_norm": 3.8850014209747314,
|
| 72 |
+
"learning_rate": 0.0009876543209876543,
|
| 73 |
+
"loss": 1.7738,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.02,
|
| 78 |
+
"grad_norm": 6.418526649475098,
|
| 79 |
+
"learning_rate": 0.0009722222222222222,
|
| 80 |
+
"loss": 1.8192,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.04,
|
| 85 |
+
"grad_norm": 6.298271179199219,
|
| 86 |
+
"learning_rate": 0.0009567901234567902,
|
| 87 |
+
"loss": 1.7833,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.05,
|
| 92 |
+
"grad_norm": 4.652337551116943,
|
| 93 |
+
"learning_rate": 0.000941358024691358,
|
| 94 |
+
"loss": 1.7534,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.07,
|
| 99 |
+
"grad_norm": 3.369626522064209,
|
| 100 |
+
"learning_rate": 0.000925925925925926,
|
| 101 |
+
"loss": 1.7376,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.08,
|
| 106 |
+
"grad_norm": 2.182107448577881,
|
| 107 |
+
"learning_rate": 0.0009104938271604939,
|
| 108 |
+
"loss": 1.6406,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.09,
|
| 113 |
+
"grad_norm": 3.763148069381714,
|
| 114 |
+
"learning_rate": 0.0008950617283950618,
|
| 115 |
+
"loss": 1.7163,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.1,
|
| 120 |
+
"eval_accuracy": 0.3373071528751753,
|
| 121 |
+
"eval_loss": 1.66623854637146,
|
| 122 |
+
"eval_runtime": 421.0248,
|
| 123 |
+
"eval_samples_per_second": 3.387,
|
| 124 |
+
"eval_steps_per_second": 0.214,
|
| 125 |
+
"step": 146
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.01,
|
| 129 |
+
"grad_norm": 4.109099864959717,
|
| 130 |
+
"learning_rate": 0.0008796296296296296,
|
| 131 |
+
"loss": 1.6734,
|
| 132 |
+
"step": 150
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.02,
|
| 136 |
+
"grad_norm": 3.7187507152557373,
|
| 137 |
+
"learning_rate": 0.0008641975308641975,
|
| 138 |
+
"loss": 1.7535,
|
| 139 |
+
"step": 160
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.03,
|
| 143 |
+
"grad_norm": 2.9596164226531982,
|
| 144 |
+
"learning_rate": 0.0008487654320987654,
|
| 145 |
+
"loss": 1.7282,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.05,
|
| 150 |
+
"grad_norm": 5.126859188079834,
|
| 151 |
+
"learning_rate": 0.0008333333333333334,
|
| 152 |
+
"loss": 1.7021,
|
| 153 |
+
"step": 180
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.06,
|
| 157 |
+
"grad_norm": 2.36877703666687,
|
| 158 |
+
"learning_rate": 0.0008179012345679012,
|
| 159 |
+
"loss": 1.6977,
|
| 160 |
+
"step": 190
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.08,
|
| 164 |
+
"grad_norm": 4.42868185043335,
|
| 165 |
+
"learning_rate": 0.0008024691358024692,
|
| 166 |
+
"loss": 1.6882,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.09,
|
| 171 |
+
"grad_norm": 3.4869942665100098,
|
| 172 |
+
"learning_rate": 0.0007870370370370372,
|
| 173 |
+
"loss": 1.7018,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.1,
|
| 178 |
+
"eval_accuracy": 0.34081346423562414,
|
| 179 |
+
"eval_loss": 1.6378456354141235,
|
| 180 |
+
"eval_runtime": 423.5487,
|
| 181 |
+
"eval_samples_per_second": 3.367,
|
| 182 |
+
"eval_steps_per_second": 0.212,
|
| 183 |
+
"step": 219
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.0,
|
| 187 |
+
"grad_norm": 3.7394754886627197,
|
| 188 |
+
"learning_rate": 0.0007716049382716049,
|
| 189 |
+
"loss": 1.7356,
|
| 190 |
+
"step": 220
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.02,
|
| 194 |
+
"grad_norm": 1.3488916158676147,
|
| 195 |
+
"learning_rate": 0.0007561728395061729,
|
| 196 |
+
"loss": 1.7024,
|
| 197 |
+
"step": 230
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.03,
|
| 201 |
+
"grad_norm": 3.196521282196045,
|
| 202 |
+
"learning_rate": 0.0007407407407407407,
|
| 203 |
+
"loss": 1.7094,
|
| 204 |
+
"step": 240
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.04,
|
| 208 |
+
"grad_norm": 3.9984352588653564,
|
| 209 |
+
"learning_rate": 0.0007253086419753087,
|
| 210 |
+
"loss": 1.6481,
|
| 211 |
+
"step": 250
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.06,
|
| 215 |
+
"grad_norm": 3.6886866092681885,
|
| 216 |
+
"learning_rate": 0.0007098765432098766,
|
| 217 |
+
"loss": 1.652,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.07,
|
| 222 |
+
"grad_norm": 2.245149850845337,
|
| 223 |
+
"learning_rate": 0.0006944444444444445,
|
| 224 |
+
"loss": 1.7479,
|
| 225 |
+
"step": 270
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.08,
|
| 229 |
+
"grad_norm": 4.62326192855835,
|
| 230 |
+
"learning_rate": 0.0006790123456790124,
|
| 231 |
+
"loss": 1.7129,
|
| 232 |
+
"step": 280
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.1,
|
| 236 |
+
"grad_norm": 4.474867343902588,
|
| 237 |
+
"learning_rate": 0.0006635802469135802,
|
| 238 |
+
"loss": 1.7334,
|
| 239 |
+
"step": 290
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.1,
|
| 243 |
+
"eval_accuracy": 0.34011220196353437,
|
| 244 |
+
"eval_loss": 1.6562532186508179,
|
| 245 |
+
"eval_runtime": 383.4418,
|
| 246 |
+
"eval_samples_per_second": 3.719,
|
| 247 |
+
"eval_steps_per_second": 0.235,
|
| 248 |
+
"step": 292
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.01,
|
| 252 |
+
"grad_norm": 1.4795947074890137,
|
| 253 |
+
"learning_rate": 0.0006481481481481481,
|
| 254 |
+
"loss": 1.7427,
|
| 255 |
+
"step": 300
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.03,
|
| 259 |
+
"grad_norm": 2.173116683959961,
|
| 260 |
+
"learning_rate": 0.0006327160493827161,
|
| 261 |
+
"loss": 1.6894,
|
| 262 |
+
"step": 310
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 4.04,
|
| 266 |
+
"grad_norm": 2.731816291809082,
|
| 267 |
+
"learning_rate": 0.0006172839506172839,
|
| 268 |
+
"loss": 1.6818,
|
| 269 |
+
"step": 320
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 4.05,
|
| 273 |
+
"grad_norm": 2.120103120803833,
|
| 274 |
+
"learning_rate": 0.0006018518518518519,
|
| 275 |
+
"loss": 1.6782,
|
| 276 |
+
"step": 330
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.07,
|
| 280 |
+
"grad_norm": 3.7916502952575684,
|
| 281 |
+
"learning_rate": 0.0005864197530864199,
|
| 282 |
+
"loss": 1.7388,
|
| 283 |
+
"step": 340
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 4.08,
|
| 287 |
+
"grad_norm": 2.852003812789917,
|
| 288 |
+
"learning_rate": 0.0005709876543209876,
|
| 289 |
+
"loss": 1.7422,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.09,
|
| 294 |
+
"grad_norm": 6.300606727600098,
|
| 295 |
+
"learning_rate": 0.0005555555555555556,
|
| 296 |
+
"loss": 1.672,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.1,
|
| 301 |
+
"eval_accuracy": 0.23983169705469845,
|
| 302 |
+
"eval_loss": 1.6567574739456177,
|
| 303 |
+
"eval_runtime": 310.9306,
|
| 304 |
+
"eval_samples_per_second": 4.586,
|
| 305 |
+
"eval_steps_per_second": 0.289,
|
| 306 |
+
"step": 365
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.01,
|
| 310 |
+
"grad_norm": 5.469892501831055,
|
| 311 |
+
"learning_rate": 0.0005401234567901234,
|
| 312 |
+
"loss": 1.6958,
|
| 313 |
+
"step": 370
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 5.02,
|
| 317 |
+
"grad_norm": 3.4026269912719727,
|
| 318 |
+
"learning_rate": 0.0005246913580246914,
|
| 319 |
+
"loss": 1.7383,
|
| 320 |
+
"step": 380
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.03,
|
| 324 |
+
"grad_norm": 2.7335870265960693,
|
| 325 |
+
"learning_rate": 0.0005092592592592593,
|
| 326 |
+
"loss": 1.6763,
|
| 327 |
+
"step": 390
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 5.05,
|
| 331 |
+
"grad_norm": 4.079995155334473,
|
| 332 |
+
"learning_rate": 0.0004938271604938272,
|
| 333 |
+
"loss": 1.693,
|
| 334 |
+
"step": 400
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 5.06,
|
| 338 |
+
"grad_norm": 2.8691930770874023,
|
| 339 |
+
"learning_rate": 0.0004783950617283951,
|
| 340 |
+
"loss": 1.7293,
|
| 341 |
+
"step": 410
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.08,
|
| 345 |
+
"grad_norm": 2.4010772705078125,
|
| 346 |
+
"learning_rate": 0.000462962962962963,
|
| 347 |
+
"loss": 1.6812,
|
| 348 |
+
"step": 420
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.09,
|
| 352 |
+
"grad_norm": 4.233634948730469,
|
| 353 |
+
"learning_rate": 0.0004475308641975309,
|
| 354 |
+
"loss": 1.7095,
|
| 355 |
+
"step": 430
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 5.1,
|
| 359 |
+
"eval_accuracy": 0.3387096774193548,
|
| 360 |
+
"eval_loss": 1.6313395500183105,
|
| 361 |
+
"eval_runtime": 306.7441,
|
| 362 |
+
"eval_samples_per_second": 4.649,
|
| 363 |
+
"eval_steps_per_second": 0.293,
|
| 364 |
+
"step": 438
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 6.0,
|
| 368 |
+
"grad_norm": 2.844386100769043,
|
| 369 |
+
"learning_rate": 0.00043209876543209873,
|
| 370 |
+
"loss": 1.6502,
|
| 371 |
+
"step": 440
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 6.02,
|
| 375 |
+
"grad_norm": 3.7761685848236084,
|
| 376 |
+
"learning_rate": 0.0004166666666666667,
|
| 377 |
+
"loss": 1.6892,
|
| 378 |
+
"step": 450
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 6.03,
|
| 382 |
+
"grad_norm": 2.8903658390045166,
|
| 383 |
+
"learning_rate": 0.0004012345679012346,
|
| 384 |
+
"loss": 1.6006,
|
| 385 |
+
"step": 460
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 6.04,
|
| 389 |
+
"grad_norm": 2.548739194869995,
|
| 390 |
+
"learning_rate": 0.00038580246913580245,
|
| 391 |
+
"loss": 1.7113,
|
| 392 |
+
"step": 470
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 6.06,
|
| 396 |
+
"grad_norm": 3.1980948448181152,
|
| 397 |
+
"learning_rate": 0.00037037037037037035,
|
| 398 |
+
"loss": 1.7259,
|
| 399 |
+
"step": 480
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 6.07,
|
| 403 |
+
"grad_norm": 3.119049549102783,
|
| 404 |
+
"learning_rate": 0.0003549382716049383,
|
| 405 |
+
"loss": 1.6634,
|
| 406 |
+
"step": 490
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 6.09,
|
| 410 |
+
"grad_norm": 3.465067148208618,
|
| 411 |
+
"learning_rate": 0.0003395061728395062,
|
| 412 |
+
"loss": 1.71,
|
| 413 |
+
"step": 500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.1,
|
| 417 |
+
"grad_norm": 1.507797122001648,
|
| 418 |
+
"learning_rate": 0.00032407407407407406,
|
| 419 |
+
"loss": 1.7119,
|
| 420 |
+
"step": 510
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 6.1,
|
| 424 |
+
"eval_accuracy": 0.34081346423562414,
|
| 425 |
+
"eval_loss": 1.6309115886688232,
|
| 426 |
+
"eval_runtime": 295.3653,
|
| 427 |
+
"eval_samples_per_second": 4.828,
|
| 428 |
+
"eval_steps_per_second": 0.305,
|
| 429 |
+
"step": 511
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 7.01,
|
| 433 |
+
"grad_norm": 1.939512848854065,
|
| 434 |
+
"learning_rate": 0.00030864197530864197,
|
| 435 |
+
"loss": 1.6486,
|
| 436 |
+
"step": 520
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 7.03,
|
| 440 |
+
"grad_norm": 2.4698586463928223,
|
| 441 |
+
"learning_rate": 0.00029320987654320993,
|
| 442 |
+
"loss": 1.6754,
|
| 443 |
+
"step": 530
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.04,
|
| 447 |
+
"grad_norm": 1.9678858518600464,
|
| 448 |
+
"learning_rate": 0.0002777777777777778,
|
| 449 |
+
"loss": 1.6025,
|
| 450 |
+
"step": 540
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.05,
|
| 454 |
+
"grad_norm": 3.6356217861175537,
|
| 455 |
+
"learning_rate": 0.0002623456790123457,
|
| 456 |
+
"loss": 1.6792,
|
| 457 |
+
"step": 550
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 7.07,
|
| 461 |
+
"grad_norm": 2.781039237976074,
|
| 462 |
+
"learning_rate": 0.0002469135802469136,
|
| 463 |
+
"loss": 1.7165,
|
| 464 |
+
"step": 560
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.08,
|
| 468 |
+
"grad_norm": 4.021714687347412,
|
| 469 |
+
"learning_rate": 0.0002314814814814815,
|
| 470 |
+
"loss": 1.6836,
|
| 471 |
+
"step": 570
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.1,
|
| 475 |
+
"grad_norm": 4.392849922180176,
|
| 476 |
+
"learning_rate": 0.00021604938271604937,
|
| 477 |
+
"loss": 1.6981,
|
| 478 |
+
"step": 580
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 7.1,
|
| 482 |
+
"eval_accuracy": 0.3288920056100982,
|
| 483 |
+
"eval_loss": 1.6518133878707886,
|
| 484 |
+
"eval_runtime": 319.5699,
|
| 485 |
+
"eval_samples_per_second": 4.462,
|
| 486 |
+
"eval_steps_per_second": 0.282,
|
| 487 |
+
"step": 584
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 8.01,
|
| 491 |
+
"grad_norm": 2.291691303253174,
|
| 492 |
+
"learning_rate": 0.0002006172839506173,
|
| 493 |
+
"loss": 1.6548,
|
| 494 |
+
"step": 590
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 8.02,
|
| 498 |
+
"grad_norm": 4.06191873550415,
|
| 499 |
+
"learning_rate": 0.00018518518518518518,
|
| 500 |
+
"loss": 1.5957,
|
| 501 |
+
"step": 600
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 8.04,
|
| 505 |
+
"grad_norm": 3.735381603240967,
|
| 506 |
+
"learning_rate": 0.0001697530864197531,
|
| 507 |
+
"loss": 1.6757,
|
| 508 |
+
"step": 610
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 8.05,
|
| 512 |
+
"grad_norm": 3.7696151733398438,
|
| 513 |
+
"learning_rate": 0.00015432098765432098,
|
| 514 |
+
"loss": 1.6349,
|
| 515 |
+
"step": 620
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 8.06,
|
| 519 |
+
"grad_norm": 2.210860013961792,
|
| 520 |
+
"learning_rate": 0.0001388888888888889,
|
| 521 |
+
"loss": 1.7561,
|
| 522 |
+
"step": 630
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 8.08,
|
| 526 |
+
"grad_norm": 2.812994956970215,
|
| 527 |
+
"learning_rate": 0.0001234567901234568,
|
| 528 |
+
"loss": 1.6331,
|
| 529 |
+
"step": 640
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 8.09,
|
| 533 |
+
"grad_norm": 3.606325626373291,
|
| 534 |
+
"learning_rate": 0.00010802469135802468,
|
| 535 |
+
"loss": 1.7066,
|
| 536 |
+
"step": 650
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 8.1,
|
| 540 |
+
"eval_accuracy": 0.33099579242636745,
|
| 541 |
+
"eval_loss": 1.6313475370407104,
|
| 542 |
+
"eval_runtime": 304.3275,
|
| 543 |
+
"eval_samples_per_second": 4.686,
|
| 544 |
+
"eval_steps_per_second": 0.296,
|
| 545 |
+
"step": 657
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 9.0,
|
| 549 |
+
"grad_norm": 2.993828296661377,
|
| 550 |
+
"learning_rate": 9.259259259259259e-05,
|
| 551 |
+
"loss": 1.6485,
|
| 552 |
+
"step": 660
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 9.02,
|
| 556 |
+
"grad_norm": 1.9592925310134888,
|
| 557 |
+
"learning_rate": 7.716049382716049e-05,
|
| 558 |
+
"loss": 1.6776,
|
| 559 |
+
"step": 670
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 9.03,
|
| 563 |
+
"grad_norm": 3.105025291442871,
|
| 564 |
+
"learning_rate": 6.17283950617284e-05,
|
| 565 |
+
"loss": 1.6466,
|
| 566 |
+
"step": 680
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 9.05,
|
| 570 |
+
"grad_norm": 3.643643856048584,
|
| 571 |
+
"learning_rate": 4.6296296296296294e-05,
|
| 572 |
+
"loss": 1.6544,
|
| 573 |
+
"step": 690
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 9.06,
|
| 577 |
+
"grad_norm": 2.5237057209014893,
|
| 578 |
+
"learning_rate": 3.08641975308642e-05,
|
| 579 |
+
"loss": 1.6616,
|
| 580 |
+
"step": 700
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 9.07,
|
| 584 |
+
"grad_norm": 3.834527015686035,
|
| 585 |
+
"learning_rate": 1.54320987654321e-05,
|
| 586 |
+
"loss": 1.6271,
|
| 587 |
+
"step": 710
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 9.09,
|
| 591 |
+
"grad_norm": 1.777999997138977,
|
| 592 |
+
"learning_rate": 0.0,
|
| 593 |
+
"loss": 1.6476,
|
| 594 |
+
"step": 720
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 9.09,
|
| 598 |
+
"eval_accuracy": 0.3288920056100982,
|
| 599 |
+
"eval_loss": 1.6337770223617554,
|
| 600 |
+
"eval_runtime": 340.2408,
|
| 601 |
+
"eval_samples_per_second": 4.191,
|
| 602 |
+
"eval_steps_per_second": 0.265,
|
| 603 |
+
"step": 720
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 9.09,
|
| 607 |
+
"step": 720,
|
| 608 |
+
"total_flos": 1.4231811343419113e+19,
|
| 609 |
+
"train_loss": 1.6995894723468357,
|
| 610 |
+
"train_runtime": 6776.3985,
|
| 611 |
+
"train_samples_per_second": 1.7,
|
| 612 |
+
"train_steps_per_second": 0.106
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 9.09,
|
| 616 |
+
"eval_accuracy": 0.34811715481171546,
|
| 617 |
+
"eval_loss": 1.703281283378601,
|
| 618 |
+
"eval_runtime": 256.932,
|
| 619 |
+
"eval_samples_per_second": 4.651,
|
| 620 |
+
"eval_steps_per_second": 0.292,
|
| 621 |
+
"step": 720
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 9.09,
|
| 625 |
+
"eval_accuracy": 0.34811715481171546,
|
| 626 |
+
"eval_loss": 1.7031110525131226,
|
| 627 |
+
"eval_runtime": 266.484,
|
| 628 |
+
"eval_samples_per_second": 4.484,
|
| 629 |
+
"eval_steps_per_second": 0.281,
|
| 630 |
+
"step": 720
|
| 631 |
+
}
|
| 632 |
+
],
|
| 633 |
+
"logging_steps": 10,
|
| 634 |
+
"max_steps": 720,
|
| 635 |
+
"num_input_tokens_seen": 0,
|
| 636 |
+
"num_train_epochs": 9223372036854775807,
|
| 637 |
+
"save_steps": 500,
|
| 638 |
+
"total_flos": 1.4231811343419113e+19,
|
| 639 |
+
"train_batch_size": 16,
|
| 640 |
+
"trial_name": null,
|
| 641 |
+
"trial_params": null
|
| 642 |
+
}
|