Commit
•
990b468
1
Parent(s):
d875331
End of training
Browse files- README.md +4 -4
- all_results.json +4 -0
- test_results.json +4 -0
- trainer_state.json +577 -0
README.md
CHANGED
|
@@ -15,11 +15,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [MCG-NJU/videomae-large](https://huggingface.co/MCG-NJU/videomae-large) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- eval_loss: 1.
|
| 19 |
- eval_accuracy: 0.5805
|
| 20 |
-
- eval_runtime:
|
| 21 |
-
- eval_samples_per_second:
|
| 22 |
-
- eval_steps_per_second: 0.
|
| 23 |
- epoch: 10.0342
|
| 24 |
- step: 618
|
| 25 |
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [MCG-NJU/videomae-large](https://huggingface.co/MCG-NJU/videomae-large) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- eval_loss: 1.7345
|
| 19 |
- eval_accuracy: 0.5805
|
| 20 |
+
- eval_runtime: 19.4619
|
| 21 |
+
- eval_samples_per_second: 10.533
|
| 22 |
+
- eval_steps_per_second: 0.668
|
| 23 |
- epoch: 10.0342
|
| 24 |
- step: 618
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.5804878048780487,
|
| 3 |
+
"eval_loss": 1.7345412969589233
|
| 4 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.5804878048780487,
|
| 3 |
+
"eval_loss": 1.7345412969589233
|
| 4 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,577 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6310679611650486,
|
| 3 |
+
"best_model_checkpoint": "videomae-large-finetuned-right-hand-conflab-v3/checkpoint-531",
|
| 4 |
+
"epoch": 10.034188034188034,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 618,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01221001221001221,
|
| 13 |
+
"grad_norm": 10.883779525756836,
|
| 14 |
+
"learning_rate": 6.0975609756097564e-06,
|
| 15 |
+
"loss": 2.1132,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02442002442002442,
|
| 20 |
+
"grad_norm": 9.386421203613281,
|
| 21 |
+
"learning_rate": 1.2195121951219513e-05,
|
| 22 |
+
"loss": 2.0109,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03663003663003663,
|
| 27 |
+
"grad_norm": 9.540677070617676,
|
| 28 |
+
"learning_rate": 1.8292682926829268e-05,
|
| 29 |
+
"loss": 1.9597,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04884004884004884,
|
| 34 |
+
"grad_norm": 8.591822624206543,
|
| 35 |
+
"learning_rate": 2.4390243902439026e-05,
|
| 36 |
+
"loss": 1.9359,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06105006105006105,
|
| 41 |
+
"grad_norm": 8.493247985839844,
|
| 42 |
+
"learning_rate": 3.048780487804878e-05,
|
| 43 |
+
"loss": 1.9038,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07203907203907203,
|
| 48 |
+
"eval_accuracy": 0.24757281553398058,
|
| 49 |
+
"eval_loss": 2.092643976211548,
|
| 50 |
+
"eval_runtime": 20.1931,
|
| 51 |
+
"eval_samples_per_second": 10.201,
|
| 52 |
+
"eval_steps_per_second": 0.644,
|
| 53 |
+
"step": 59
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.0012210012210012,
|
| 57 |
+
"grad_norm": 9.048627853393555,
|
| 58 |
+
"learning_rate": 3.6585365853658535e-05,
|
| 59 |
+
"loss": 2.0497,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.0134310134310134,
|
| 64 |
+
"grad_norm": 5.44881010055542,
|
| 65 |
+
"learning_rate": 4.26829268292683e-05,
|
| 66 |
+
"loss": 2.0479,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.0256410256410255,
|
| 71 |
+
"grad_norm": 6.384305477142334,
|
| 72 |
+
"learning_rate": 4.878048780487805e-05,
|
| 73 |
+
"loss": 2.09,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.037851037851038,
|
| 78 |
+
"grad_norm": 4.374388694763184,
|
| 79 |
+
"learning_rate": 4.94572591587517e-05,
|
| 80 |
+
"loss": 2.0209,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.05006105006105,
|
| 85 |
+
"grad_norm": 4.627490043640137,
|
| 86 |
+
"learning_rate": 4.877883310719132e-05,
|
| 87 |
+
"loss": 2.0525,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.0622710622710623,
|
| 92 |
+
"grad_norm": 5.586881160736084,
|
| 93 |
+
"learning_rate": 4.810040705563094e-05,
|
| 94 |
+
"loss": 1.8963,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.072039072039072,
|
| 99 |
+
"eval_accuracy": 0.24271844660194175,
|
| 100 |
+
"eval_loss": 1.9697831869125366,
|
| 101 |
+
"eval_runtime": 21.1947,
|
| 102 |
+
"eval_samples_per_second": 9.719,
|
| 103 |
+
"eval_steps_per_second": 0.613,
|
| 104 |
+
"step": 118
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.0024420024420024,
|
| 108 |
+
"grad_norm": 5.62977409362793,
|
| 109 |
+
"learning_rate": 4.742198100407056e-05,
|
| 110 |
+
"loss": 1.9983,
|
| 111 |
+
"step": 120
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.0146520146520146,
|
| 115 |
+
"grad_norm": 4.484747886657715,
|
| 116 |
+
"learning_rate": 4.674355495251018e-05,
|
| 117 |
+
"loss": 1.9586,
|
| 118 |
+
"step": 130
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.0268620268620268,
|
| 122 |
+
"grad_norm": 4.371191501617432,
|
| 123 |
+
"learning_rate": 4.60651289009498e-05,
|
| 124 |
+
"loss": 1.8729,
|
| 125 |
+
"step": 140
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.039072039072039,
|
| 129 |
+
"grad_norm": 4.68417501449585,
|
| 130 |
+
"learning_rate": 4.5386702849389416e-05,
|
| 131 |
+
"loss": 2.0021,
|
| 132 |
+
"step": 150
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.051282051282051,
|
| 136 |
+
"grad_norm": 12.167759895324707,
|
| 137 |
+
"learning_rate": 4.470827679782904e-05,
|
| 138 |
+
"loss": 1.7877,
|
| 139 |
+
"step": 160
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.0634920634920633,
|
| 143 |
+
"grad_norm": 3.787353515625,
|
| 144 |
+
"learning_rate": 4.402985074626866e-05,
|
| 145 |
+
"loss": 1.9154,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.0720390720390722,
|
| 150 |
+
"eval_accuracy": 0.36893203883495146,
|
| 151 |
+
"eval_loss": 1.7126853466033936,
|
| 152 |
+
"eval_runtime": 18.7207,
|
| 153 |
+
"eval_samples_per_second": 11.004,
|
| 154 |
+
"eval_steps_per_second": 0.694,
|
| 155 |
+
"step": 177
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.0036630036630036,
|
| 159 |
+
"grad_norm": 5.718711853027344,
|
| 160 |
+
"learning_rate": 4.335142469470828e-05,
|
| 161 |
+
"loss": 1.7051,
|
| 162 |
+
"step": 180
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.015873015873016,
|
| 166 |
+
"grad_norm": 6.827328681945801,
|
| 167 |
+
"learning_rate": 4.26729986431479e-05,
|
| 168 |
+
"loss": 1.7233,
|
| 169 |
+
"step": 190
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 3.028083028083028,
|
| 173 |
+
"grad_norm": 6.945235252380371,
|
| 174 |
+
"learning_rate": 4.199457259158752e-05,
|
| 175 |
+
"loss": 1.5634,
|
| 176 |
+
"step": 200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.04029304029304,
|
| 180 |
+
"grad_norm": 7.2410197257995605,
|
| 181 |
+
"learning_rate": 4.131614654002714e-05,
|
| 182 |
+
"loss": 1.5571,
|
| 183 |
+
"step": 210
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.0525030525030523,
|
| 187 |
+
"grad_norm": 5.356701374053955,
|
| 188 |
+
"learning_rate": 4.063772048846676e-05,
|
| 189 |
+
"loss": 1.6352,
|
| 190 |
+
"step": 220
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.064713064713065,
|
| 194 |
+
"grad_norm": 6.82385778427124,
|
| 195 |
+
"learning_rate": 3.995929443690638e-05,
|
| 196 |
+
"loss": 1.4173,
|
| 197 |
+
"step": 230
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.0720390720390722,
|
| 201 |
+
"eval_accuracy": 0.5242718446601942,
|
| 202 |
+
"eval_loss": 1.4571937322616577,
|
| 203 |
+
"eval_runtime": 25.9349,
|
| 204 |
+
"eval_samples_per_second": 7.943,
|
| 205 |
+
"eval_steps_per_second": 0.501,
|
| 206 |
+
"step": 236
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.004884004884005,
|
| 210 |
+
"grad_norm": 5.349239349365234,
|
| 211 |
+
"learning_rate": 3.9280868385345995e-05,
|
| 212 |
+
"loss": 1.5368,
|
| 213 |
+
"step": 240
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 4.017094017094017,
|
| 217 |
+
"grad_norm": 7.013088703155518,
|
| 218 |
+
"learning_rate": 3.860244233378562e-05,
|
| 219 |
+
"loss": 1.3767,
|
| 220 |
+
"step": 250
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 4.029304029304029,
|
| 224 |
+
"grad_norm": 8.280290603637695,
|
| 225 |
+
"learning_rate": 3.792401628222524e-05,
|
| 226 |
+
"loss": 1.4361,
|
| 227 |
+
"step": 260
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 4.041514041514041,
|
| 231 |
+
"grad_norm": 8.19866943359375,
|
| 232 |
+
"learning_rate": 3.724559023066486e-05,
|
| 233 |
+
"loss": 1.193,
|
| 234 |
+
"step": 270
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.0537240537240535,
|
| 238 |
+
"grad_norm": 7.372025012969971,
|
| 239 |
+
"learning_rate": 3.656716417910448e-05,
|
| 240 |
+
"loss": 1.3667,
|
| 241 |
+
"step": 280
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 4.065934065934066,
|
| 245 |
+
"grad_norm": 5.639531135559082,
|
| 246 |
+
"learning_rate": 3.58887381275441e-05,
|
| 247 |
+
"loss": 1.3246,
|
| 248 |
+
"step": 290
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.072039072039072,
|
| 252 |
+
"eval_accuracy": 0.5679611650485437,
|
| 253 |
+
"eval_loss": 1.223657488822937,
|
| 254 |
+
"eval_runtime": 20.7954,
|
| 255 |
+
"eval_samples_per_second": 9.906,
|
| 256 |
+
"eval_steps_per_second": 0.625,
|
| 257 |
+
"step": 295
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 5.006105006105006,
|
| 261 |
+
"grad_norm": 7.878034591674805,
|
| 262 |
+
"learning_rate": 3.521031207598372e-05,
|
| 263 |
+
"loss": 1.1881,
|
| 264 |
+
"step": 300
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 5.018315018315018,
|
| 268 |
+
"grad_norm": 7.679462909698486,
|
| 269 |
+
"learning_rate": 3.453188602442334e-05,
|
| 270 |
+
"loss": 1.1117,
|
| 271 |
+
"step": 310
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 5.03052503052503,
|
| 275 |
+
"grad_norm": 10.07403564453125,
|
| 276 |
+
"learning_rate": 3.385345997286296e-05,
|
| 277 |
+
"loss": 1.0766,
|
| 278 |
+
"step": 320
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.042735042735043,
|
| 282 |
+
"grad_norm": 7.193727970123291,
|
| 283 |
+
"learning_rate": 3.3175033921302575e-05,
|
| 284 |
+
"loss": 1.006,
|
| 285 |
+
"step": 330
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 5.054945054945055,
|
| 289 |
+
"grad_norm": 5.925375461578369,
|
| 290 |
+
"learning_rate": 3.24966078697422e-05,
|
| 291 |
+
"loss": 0.9709,
|
| 292 |
+
"step": 340
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 5.067155067155067,
|
| 296 |
+
"grad_norm": 13.94832706451416,
|
| 297 |
+
"learning_rate": 3.181818181818182e-05,
|
| 298 |
+
"loss": 1.077,
|
| 299 |
+
"step": 350
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 5.072039072039072,
|
| 303 |
+
"eval_accuracy": 0.49514563106796117,
|
| 304 |
+
"eval_loss": 1.4524530172348022,
|
| 305 |
+
"eval_runtime": 22.2248,
|
| 306 |
+
"eval_samples_per_second": 9.269,
|
| 307 |
+
"eval_steps_per_second": 0.585,
|
| 308 |
+
"step": 354
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.007326007326007,
|
| 312 |
+
"grad_norm": 8.016749382019043,
|
| 313 |
+
"learning_rate": 3.113975576662144e-05,
|
| 314 |
+
"loss": 0.9269,
|
| 315 |
+
"step": 360
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 6.0195360195360195,
|
| 319 |
+
"grad_norm": 5.849711894989014,
|
| 320 |
+
"learning_rate": 3.046132971506106e-05,
|
| 321 |
+
"loss": 0.8371,
|
| 322 |
+
"step": 370
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 6.031746031746032,
|
| 326 |
+
"grad_norm": 7.203096389770508,
|
| 327 |
+
"learning_rate": 2.9782903663500678e-05,
|
| 328 |
+
"loss": 0.7639,
|
| 329 |
+
"step": 380
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 6.043956043956044,
|
| 333 |
+
"grad_norm": 7.49542236328125,
|
| 334 |
+
"learning_rate": 2.91044776119403e-05,
|
| 335 |
+
"loss": 0.8475,
|
| 336 |
+
"step": 390
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 6.056166056166056,
|
| 340 |
+
"grad_norm": 9.251474380493164,
|
| 341 |
+
"learning_rate": 2.842605156037992e-05,
|
| 342 |
+
"loss": 0.8942,
|
| 343 |
+
"step": 400
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 6.068376068376068,
|
| 347 |
+
"grad_norm": 8.035192489624023,
|
| 348 |
+
"learning_rate": 2.7747625508819542e-05,
|
| 349 |
+
"loss": 0.923,
|
| 350 |
+
"step": 410
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 6.072039072039072,
|
| 354 |
+
"eval_accuracy": 0.587378640776699,
|
| 355 |
+
"eval_loss": 1.3191975355148315,
|
| 356 |
+
"eval_runtime": 21.2599,
|
| 357 |
+
"eval_samples_per_second": 9.69,
|
| 358 |
+
"eval_steps_per_second": 0.611,
|
| 359 |
+
"step": 413
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 7.0085470085470085,
|
| 363 |
+
"grad_norm": 9.951005935668945,
|
| 364 |
+
"learning_rate": 2.7069199457259158e-05,
|
| 365 |
+
"loss": 0.6643,
|
| 366 |
+
"step": 420
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 7.020757020757021,
|
| 370 |
+
"grad_norm": 10.067875862121582,
|
| 371 |
+
"learning_rate": 2.639077340569878e-05,
|
| 372 |
+
"loss": 0.6866,
|
| 373 |
+
"step": 430
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 7.032967032967033,
|
| 377 |
+
"grad_norm": 11.081637382507324,
|
| 378 |
+
"learning_rate": 2.57123473541384e-05,
|
| 379 |
+
"loss": 0.5887,
|
| 380 |
+
"step": 440
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 7.045177045177045,
|
| 384 |
+
"grad_norm": 7.662527084350586,
|
| 385 |
+
"learning_rate": 2.5033921302578023e-05,
|
| 386 |
+
"loss": 0.6175,
|
| 387 |
+
"step": 450
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.057387057387057,
|
| 391 |
+
"grad_norm": 6.395303726196289,
|
| 392 |
+
"learning_rate": 2.4355495251017642e-05,
|
| 393 |
+
"loss": 0.6575,
|
| 394 |
+
"step": 460
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 7.069597069597069,
|
| 398 |
+
"grad_norm": 12.816076278686523,
|
| 399 |
+
"learning_rate": 2.367706919945726e-05,
|
| 400 |
+
"loss": 0.651,
|
| 401 |
+
"step": 470
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.072039072039072,
|
| 405 |
+
"eval_accuracy": 0.5728155339805825,
|
| 406 |
+
"eval_loss": 1.3817497491836548,
|
| 407 |
+
"eval_runtime": 22.418,
|
| 408 |
+
"eval_samples_per_second": 9.189,
|
| 409 |
+
"eval_steps_per_second": 0.58,
|
| 410 |
+
"step": 472
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 8.00976800976801,
|
| 414 |
+
"grad_norm": 6.377563953399658,
|
| 415 |
+
"learning_rate": 2.299864314789688e-05,
|
| 416 |
+
"loss": 0.4579,
|
| 417 |
+
"step": 480
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 8.021978021978022,
|
| 421 |
+
"grad_norm": 9.152300834655762,
|
| 422 |
+
"learning_rate": 2.2320217096336503e-05,
|
| 423 |
+
"loss": 0.4142,
|
| 424 |
+
"step": 490
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 8.034188034188034,
|
| 428 |
+
"grad_norm": 6.605992317199707,
|
| 429 |
+
"learning_rate": 2.164179104477612e-05,
|
| 430 |
+
"loss": 0.4407,
|
| 431 |
+
"step": 500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 8.046398046398046,
|
| 435 |
+
"grad_norm": 14.947940826416016,
|
| 436 |
+
"learning_rate": 2.0963364993215738e-05,
|
| 437 |
+
"loss": 0.3638,
|
| 438 |
+
"step": 510
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 8.058608058608058,
|
| 442 |
+
"grad_norm": 9.7440824508667,
|
| 443 |
+
"learning_rate": 2.028493894165536e-05,
|
| 444 |
+
"loss": 0.5602,
|
| 445 |
+
"step": 520
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 8.07081807081807,
|
| 449 |
+
"grad_norm": 8.176590919494629,
|
| 450 |
+
"learning_rate": 1.960651289009498e-05,
|
| 451 |
+
"loss": 0.5092,
|
| 452 |
+
"step": 530
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 8.072039072039072,
|
| 456 |
+
"eval_accuracy": 0.6310679611650486,
|
| 457 |
+
"eval_loss": 1.3680421113967896,
|
| 458 |
+
"eval_runtime": 18.341,
|
| 459 |
+
"eval_samples_per_second": 11.232,
|
| 460 |
+
"eval_steps_per_second": 0.709,
|
| 461 |
+
"step": 531
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 9.010989010989011,
|
| 465 |
+
"grad_norm": 6.507408142089844,
|
| 466 |
+
"learning_rate": 1.89280868385346e-05,
|
| 467 |
+
"loss": 0.3515,
|
| 468 |
+
"step": 540
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 9.023199023199023,
|
| 472 |
+
"grad_norm": 10.037909507751465,
|
| 473 |
+
"learning_rate": 1.824966078697422e-05,
|
| 474 |
+
"loss": 0.38,
|
| 475 |
+
"step": 550
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 9.035409035409035,
|
| 479 |
+
"grad_norm": 8.945121765136719,
|
| 480 |
+
"learning_rate": 1.757123473541384e-05,
|
| 481 |
+
"loss": 0.2686,
|
| 482 |
+
"step": 560
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 9.047619047619047,
|
| 486 |
+
"grad_norm": 7.725048542022705,
|
| 487 |
+
"learning_rate": 1.689280868385346e-05,
|
| 488 |
+
"loss": 0.318,
|
| 489 |
+
"step": 570
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 9.05982905982906,
|
| 493 |
+
"grad_norm": 8.256610870361328,
|
| 494 |
+
"learning_rate": 1.6214382632293083e-05,
|
| 495 |
+
"loss": 0.3725,
|
| 496 |
+
"step": 580
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 9.072039072039072,
|
| 500 |
+
"grad_norm": 11.454293251037598,
|
| 501 |
+
"learning_rate": 1.55359565807327e-05,
|
| 502 |
+
"loss": 0.2484,
|
| 503 |
+
"step": 590
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 9.072039072039072,
|
| 507 |
+
"eval_accuracy": 0.6262135922330098,
|
| 508 |
+
"eval_loss": 1.4326366186141968,
|
| 509 |
+
"eval_runtime": 19.1344,
|
| 510 |
+
"eval_samples_per_second": 10.766,
|
| 511 |
+
"eval_steps_per_second": 0.679,
|
| 512 |
+
"step": 590
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 10.012210012210012,
|
| 516 |
+
"grad_norm": 16.326658248901367,
|
| 517 |
+
"learning_rate": 1.485753052917232e-05,
|
| 518 |
+
"loss": 0.2977,
|
| 519 |
+
"step": 600
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 10.024420024420024,
|
| 523 |
+
"grad_norm": 6.485968112945557,
|
| 524 |
+
"learning_rate": 1.417910447761194e-05,
|
| 525 |
+
"loss": 0.2218,
|
| 526 |
+
"step": 610
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 10.034188034188034,
|
| 530 |
+
"eval_accuracy": 0.5804878048780487,
|
| 531 |
+
"eval_loss": 1.7306667566299438,
|
| 532 |
+
"eval_runtime": 28.969,
|
| 533 |
+
"eval_samples_per_second": 7.077,
|
| 534 |
+
"eval_steps_per_second": 0.449,
|
| 535 |
+
"step": 618
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 10.034188034188034,
|
| 539 |
+
"eval_accuracy": 0.5804878048780487,
|
| 540 |
+
"eval_loss": 1.7345412969589233,
|
| 541 |
+
"eval_runtime": 19.4619,
|
| 542 |
+
"eval_samples_per_second": 10.533,
|
| 543 |
+
"eval_steps_per_second": 0.668,
|
| 544 |
+
"step": 618
|
| 545 |
+
}
|
| 546 |
+
],
|
| 547 |
+
"logging_steps": 10,
|
| 548 |
+
"max_steps": 819,
|
| 549 |
+
"num_input_tokens_seen": 0,
|
| 550 |
+
"num_train_epochs": 9223372036854775807,
|
| 551 |
+
"save_steps": 500,
|
| 552 |
+
"stateful_callbacks": {
|
| 553 |
+
"EarlyStoppingCallback": {
|
| 554 |
+
"args": {
|
| 555 |
+
"early_stopping_patience": 3,
|
| 556 |
+
"early_stopping_threshold": 0.0
|
| 557 |
+
},
|
| 558 |
+
"attributes": {
|
| 559 |
+
"early_stopping_patience_counter": 0
|
| 560 |
+
}
|
| 561 |
+
},
|
| 562 |
+
"TrainerControl": {
|
| 563 |
+
"args": {
|
| 564 |
+
"should_epoch_stop": false,
|
| 565 |
+
"should_evaluate": false,
|
| 566 |
+
"should_log": false,
|
| 567 |
+
"should_save": true,
|
| 568 |
+
"should_training_stop": false
|
| 569 |
+
},
|
| 570 |
+
"attributes": {}
|
| 571 |
+
}
|
| 572 |
+
},
|
| 573 |
+
"total_flos": 4.276946192165241e+19,
|
| 574 |
+
"train_batch_size": 16,
|
| 575 |
+
"trial_name": null,
|
| 576 |
+
"trial_params": null
|
| 577 |
+
}
|