defect
Browse files- all_results.json +18 -0
- eval_results.json +8 -0
- runs/Jun14_00-20-54_e674c9136a07/events.out.tfevents.1686738783.e674c9136a07.10000.1 +3 -0
- test_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +700 -0
all_results.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.7542857142857143,
|
| 4 |
+
"eval_loss": 0.7207300066947937,
|
| 5 |
+
"eval_runtime": 226.5026,
|
| 6 |
+
"eval_samples_per_second": 1.545,
|
| 7 |
+
"eval_steps_per_second": 0.194,
|
| 8 |
+
"test_accuracy": 0.7182320441988951,
|
| 9 |
+
"test_loss": 0.7332790493965149,
|
| 10 |
+
"test_runtime": 164.6516,
|
| 11 |
+
"test_samples_per_second": 1.099,
|
| 12 |
+
"test_steps_per_second": 0.14,
|
| 13 |
+
"total_flos": 9.8805828893184e+17,
|
| 14 |
+
"train_loss": 0.09906705205639203,
|
| 15 |
+
"train_runtime": 36217.9644,
|
| 16 |
+
"train_samples_per_second": 0.352,
|
| 17 |
+
"train_steps_per_second": 0.012
|
| 18 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.7542857142857143,
|
| 4 |
+
"eval_loss": 0.7207300066947937,
|
| 5 |
+
"eval_runtime": 226.5026,
|
| 6 |
+
"eval_samples_per_second": 1.545,
|
| 7 |
+
"eval_steps_per_second": 0.194
|
| 8 |
+
}
|
runs/Jun14_00-20-54_e674c9136a07/events.out.tfevents.1686738783.e674c9136a07.10000.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8bff1a512d8d07675c0a067964f6abb306b53bad026626d5d25908cbbf53cf1a
|
| 3 |
+
size 411
|
test_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"test_accuracy": 0.7182320441988951,
|
| 3 |
+
"test_loss": 0.7332790493965149,
|
| 4 |
+
"test_runtime": 164.6516,
|
| 5 |
+
"test_samples_per_second": 1.099,
|
| 6 |
+
"test_steps_per_second": 0.14
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"total_flos": 9.8805828893184e+17,
|
| 4 |
+
"train_loss": 0.09906705205639203,
|
| 5 |
+
"train_runtime": 36217.9644,
|
| 6 |
+
"train_samples_per_second": 0.352,
|
| 7 |
+
"train_steps_per_second": 0.012
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,700 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7207300066947937,
|
| 3 |
+
"best_model_checkpoint": "/raildefectfft2/checkpoint-30",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"global_step": 450,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.67,
|
| 12 |
+
"learning_rate": 0.00019555555555555556,
|
| 13 |
+
"loss": 1.3922,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.67,
|
| 18 |
+
"eval_accuracy": 0.6114285714285714,
|
| 19 |
+
"eval_loss": 1.1690133810043335,
|
| 20 |
+
"eval_runtime": 217.3963,
|
| 21 |
+
"eval_samples_per_second": 1.61,
|
| 22 |
+
"eval_steps_per_second": 0.202,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.33,
|
| 27 |
+
"learning_rate": 0.00019111111111111114,
|
| 28 |
+
"loss": 0.8518,
|
| 29 |
+
"step": 20
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 1.33,
|
| 33 |
+
"eval_accuracy": 0.6828571428571428,
|
| 34 |
+
"eval_loss": 0.8874489068984985,
|
| 35 |
+
"eval_runtime": 217.0024,
|
| 36 |
+
"eval_samples_per_second": 1.613,
|
| 37 |
+
"eval_steps_per_second": 0.203,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.0,
|
| 42 |
+
"learning_rate": 0.0001866666666666667,
|
| 43 |
+
"loss": 0.5386,
|
| 44 |
+
"step": 30
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 2.0,
|
| 48 |
+
"eval_accuracy": 0.7542857142857143,
|
| 49 |
+
"eval_loss": 0.7207300066947937,
|
| 50 |
+
"eval_runtime": 221.2829,
|
| 51 |
+
"eval_samples_per_second": 1.582,
|
| 52 |
+
"eval_steps_per_second": 0.199,
|
| 53 |
+
"step": 30
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.67,
|
| 57 |
+
"learning_rate": 0.00018222222222222224,
|
| 58 |
+
"loss": 0.3125,
|
| 59 |
+
"step": 40
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.67,
|
| 63 |
+
"eval_accuracy": 0.7285714285714285,
|
| 64 |
+
"eval_loss": 0.8382583260536194,
|
| 65 |
+
"eval_runtime": 217.2612,
|
| 66 |
+
"eval_samples_per_second": 1.611,
|
| 67 |
+
"eval_steps_per_second": 0.203,
|
| 68 |
+
"step": 40
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.33,
|
| 72 |
+
"learning_rate": 0.00017777777777777779,
|
| 73 |
+
"loss": 0.2264,
|
| 74 |
+
"step": 50
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 3.33,
|
| 78 |
+
"eval_accuracy": 0.7428571428571429,
|
| 79 |
+
"eval_loss": 0.8440293669700623,
|
| 80 |
+
"eval_runtime": 215.3358,
|
| 81 |
+
"eval_samples_per_second": 1.625,
|
| 82 |
+
"eval_steps_per_second": 0.204,
|
| 83 |
+
"step": 50
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 4.0,
|
| 87 |
+
"learning_rate": 0.00017333333333333334,
|
| 88 |
+
"loss": 0.1613,
|
| 89 |
+
"step": 60
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 4.0,
|
| 93 |
+
"eval_accuracy": 0.7457142857142857,
|
| 94 |
+
"eval_loss": 0.8516280055046082,
|
| 95 |
+
"eval_runtime": 221.3905,
|
| 96 |
+
"eval_samples_per_second": 1.581,
|
| 97 |
+
"eval_steps_per_second": 0.199,
|
| 98 |
+
"step": 60
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 4.67,
|
| 102 |
+
"learning_rate": 0.00016888888888888889,
|
| 103 |
+
"loss": 0.119,
|
| 104 |
+
"step": 70
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.67,
|
| 108 |
+
"eval_accuracy": 0.6,
|
| 109 |
+
"eval_loss": 1.3625210523605347,
|
| 110 |
+
"eval_runtime": 210.1415,
|
| 111 |
+
"eval_samples_per_second": 1.666,
|
| 112 |
+
"eval_steps_per_second": 0.209,
|
| 113 |
+
"step": 70
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 5.33,
|
| 117 |
+
"learning_rate": 0.00016444444444444444,
|
| 118 |
+
"loss": 0.0972,
|
| 119 |
+
"step": 80
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 5.33,
|
| 123 |
+
"eval_accuracy": 0.7428571428571429,
|
| 124 |
+
"eval_loss": 0.9109606146812439,
|
| 125 |
+
"eval_runtime": 208.8901,
|
| 126 |
+
"eval_samples_per_second": 1.676,
|
| 127 |
+
"eval_steps_per_second": 0.211,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 6.0,
|
| 132 |
+
"learning_rate": 0.00016,
|
| 133 |
+
"loss": 0.0844,
|
| 134 |
+
"step": 90
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"eval_accuracy": 0.78,
|
| 139 |
+
"eval_loss": 0.8271930813789368,
|
| 140 |
+
"eval_runtime": 214.6994,
|
| 141 |
+
"eval_samples_per_second": 1.63,
|
| 142 |
+
"eval_steps_per_second": 0.205,
|
| 143 |
+
"step": 90
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.67,
|
| 147 |
+
"learning_rate": 0.00015555555555555556,
|
| 148 |
+
"loss": 0.0725,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 6.67,
|
| 153 |
+
"eval_accuracy": 0.74,
|
| 154 |
+
"eval_loss": 0.8958377242088318,
|
| 155 |
+
"eval_runtime": 209.0619,
|
| 156 |
+
"eval_samples_per_second": 1.674,
|
| 157 |
+
"eval_steps_per_second": 0.21,
|
| 158 |
+
"step": 100
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 7.33,
|
| 162 |
+
"learning_rate": 0.0001511111111111111,
|
| 163 |
+
"loss": 0.0708,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 7.33,
|
| 168 |
+
"eval_accuracy": 0.7371428571428571,
|
| 169 |
+
"eval_loss": 1.0972360372543335,
|
| 170 |
+
"eval_runtime": 210.819,
|
| 171 |
+
"eval_samples_per_second": 1.66,
|
| 172 |
+
"eval_steps_per_second": 0.209,
|
| 173 |
+
"step": 110
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 8.0,
|
| 177 |
+
"learning_rate": 0.00014666666666666666,
|
| 178 |
+
"loss": 0.041,
|
| 179 |
+
"step": 120
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 8.0,
|
| 183 |
+
"eval_accuracy": 0.7628571428571429,
|
| 184 |
+
"eval_loss": 1.0088900327682495,
|
| 185 |
+
"eval_runtime": 209.2911,
|
| 186 |
+
"eval_samples_per_second": 1.672,
|
| 187 |
+
"eval_steps_per_second": 0.21,
|
| 188 |
+
"step": 120
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 8.67,
|
| 192 |
+
"learning_rate": 0.00014222222222222224,
|
| 193 |
+
"loss": 0.0312,
|
| 194 |
+
"step": 130
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 8.67,
|
| 198 |
+
"eval_accuracy": 0.7628571428571429,
|
| 199 |
+
"eval_loss": 1.03481125831604,
|
| 200 |
+
"eval_runtime": 215.9471,
|
| 201 |
+
"eval_samples_per_second": 1.621,
|
| 202 |
+
"eval_steps_per_second": 0.204,
|
| 203 |
+
"step": 130
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 9.33,
|
| 207 |
+
"learning_rate": 0.0001377777777777778,
|
| 208 |
+
"loss": 0.0401,
|
| 209 |
+
"step": 140
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.33,
|
| 213 |
+
"eval_accuracy": 0.7257142857142858,
|
| 214 |
+
"eval_loss": 1.2426719665527344,
|
| 215 |
+
"eval_runtime": 211.184,
|
| 216 |
+
"eval_samples_per_second": 1.657,
|
| 217 |
+
"eval_steps_per_second": 0.208,
|
| 218 |
+
"step": 140
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 10.0,
|
| 222 |
+
"learning_rate": 0.00013333333333333334,
|
| 223 |
+
"loss": 0.0271,
|
| 224 |
+
"step": 150
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 10.0,
|
| 228 |
+
"eval_accuracy": 0.7542857142857143,
|
| 229 |
+
"eval_loss": 1.0153539180755615,
|
| 230 |
+
"eval_runtime": 208.7196,
|
| 231 |
+
"eval_samples_per_second": 1.677,
|
| 232 |
+
"eval_steps_per_second": 0.211,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 10.67,
|
| 237 |
+
"learning_rate": 0.00012888888888888892,
|
| 238 |
+
"loss": 0.0328,
|
| 239 |
+
"step": 160
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 10.67,
|
| 243 |
+
"eval_accuracy": 0.7714285714285715,
|
| 244 |
+
"eval_loss": 1.0373210906982422,
|
| 245 |
+
"eval_runtime": 216.803,
|
| 246 |
+
"eval_samples_per_second": 1.614,
|
| 247 |
+
"eval_steps_per_second": 0.203,
|
| 248 |
+
"step": 160
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 11.33,
|
| 252 |
+
"learning_rate": 0.00012444444444444444,
|
| 253 |
+
"loss": 0.023,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 11.33,
|
| 258 |
+
"eval_accuracy": 0.7685714285714286,
|
| 259 |
+
"eval_loss": 1.005110502243042,
|
| 260 |
+
"eval_runtime": 209.183,
|
| 261 |
+
"eval_samples_per_second": 1.673,
|
| 262 |
+
"eval_steps_per_second": 0.21,
|
| 263 |
+
"step": 170
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 12.0,
|
| 267 |
+
"learning_rate": 0.00012,
|
| 268 |
+
"loss": 0.0199,
|
| 269 |
+
"step": 180
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 12.0,
|
| 273 |
+
"eval_accuracy": 0.7657142857142857,
|
| 274 |
+
"eval_loss": 0.9775477647781372,
|
| 275 |
+
"eval_runtime": 208.6639,
|
| 276 |
+
"eval_samples_per_second": 1.677,
|
| 277 |
+
"eval_steps_per_second": 0.211,
|
| 278 |
+
"step": 180
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 12.67,
|
| 282 |
+
"learning_rate": 0.00011555555555555555,
|
| 283 |
+
"loss": 0.0189,
|
| 284 |
+
"step": 190
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 12.67,
|
| 288 |
+
"eval_accuracy": 0.7657142857142857,
|
| 289 |
+
"eval_loss": 1.008815050125122,
|
| 290 |
+
"eval_runtime": 216.3653,
|
| 291 |
+
"eval_samples_per_second": 1.618,
|
| 292 |
+
"eval_steps_per_second": 0.203,
|
| 293 |
+
"step": 190
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 13.33,
|
| 297 |
+
"learning_rate": 0.00011111111111111112,
|
| 298 |
+
"loss": 0.0188,
|
| 299 |
+
"step": 200
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 13.33,
|
| 303 |
+
"eval_accuracy": 0.7342857142857143,
|
| 304 |
+
"eval_loss": 1.1904319524765015,
|
| 305 |
+
"eval_runtime": 209.8186,
|
| 306 |
+
"eval_samples_per_second": 1.668,
|
| 307 |
+
"eval_steps_per_second": 0.21,
|
| 308 |
+
"step": 200
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 14.0,
|
| 312 |
+
"learning_rate": 0.00010666666666666667,
|
| 313 |
+
"loss": 0.0167,
|
| 314 |
+
"step": 210
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 14.0,
|
| 318 |
+
"eval_accuracy": 0.7285714285714285,
|
| 319 |
+
"eval_loss": 1.2999135255813599,
|
| 320 |
+
"eval_runtime": 209.7587,
|
| 321 |
+
"eval_samples_per_second": 1.669,
|
| 322 |
+
"eval_steps_per_second": 0.21,
|
| 323 |
+
"step": 210
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 14.67,
|
| 327 |
+
"learning_rate": 0.00010222222222222222,
|
| 328 |
+
"loss": 0.0159,
|
| 329 |
+
"step": 220
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 14.67,
|
| 333 |
+
"eval_accuracy": 0.7514285714285714,
|
| 334 |
+
"eval_loss": 1.1326370239257812,
|
| 335 |
+
"eval_runtime": 215.8574,
|
| 336 |
+
"eval_samples_per_second": 1.621,
|
| 337 |
+
"eval_steps_per_second": 0.204,
|
| 338 |
+
"step": 220
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 15.33,
|
| 342 |
+
"learning_rate": 9.777777777777778e-05,
|
| 343 |
+
"loss": 0.0145,
|
| 344 |
+
"step": 230
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 15.33,
|
| 348 |
+
"eval_accuracy": 0.7542857142857143,
|
| 349 |
+
"eval_loss": 1.1385791301727295,
|
| 350 |
+
"eval_runtime": 209.3052,
|
| 351 |
+
"eval_samples_per_second": 1.672,
|
| 352 |
+
"eval_steps_per_second": 0.21,
|
| 353 |
+
"step": 230
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 16.0,
|
| 357 |
+
"learning_rate": 9.333333333333334e-05,
|
| 358 |
+
"loss": 0.015,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 16.0,
|
| 363 |
+
"eval_accuracy": 0.7542857142857143,
|
| 364 |
+
"eval_loss": 1.1441489458084106,
|
| 365 |
+
"eval_runtime": 209.1403,
|
| 366 |
+
"eval_samples_per_second": 1.674,
|
| 367 |
+
"eval_steps_per_second": 0.21,
|
| 368 |
+
"step": 240
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 16.67,
|
| 372 |
+
"learning_rate": 8.888888888888889e-05,
|
| 373 |
+
"loss": 0.0133,
|
| 374 |
+
"step": 250
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 16.67,
|
| 378 |
+
"eval_accuracy": 0.7514285714285714,
|
| 379 |
+
"eval_loss": 1.154445767402649,
|
| 380 |
+
"eval_runtime": 212.9184,
|
| 381 |
+
"eval_samples_per_second": 1.644,
|
| 382 |
+
"eval_steps_per_second": 0.207,
|
| 383 |
+
"step": 250
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 17.33,
|
| 387 |
+
"learning_rate": 8.444444444444444e-05,
|
| 388 |
+
"loss": 0.0132,
|
| 389 |
+
"step": 260
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 17.33,
|
| 393 |
+
"eval_accuracy": 0.7514285714285714,
|
| 394 |
+
"eval_loss": 1.1629431247711182,
|
| 395 |
+
"eval_runtime": 209.529,
|
| 396 |
+
"eval_samples_per_second": 1.67,
|
| 397 |
+
"eval_steps_per_second": 0.21,
|
| 398 |
+
"step": 260
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 18.0,
|
| 402 |
+
"learning_rate": 8e-05,
|
| 403 |
+
"loss": 0.0121,
|
| 404 |
+
"step": 270
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 18.0,
|
| 408 |
+
"eval_accuracy": 0.7514285714285714,
|
| 409 |
+
"eval_loss": 1.1707779169082642,
|
| 410 |
+
"eval_runtime": 209.7822,
|
| 411 |
+
"eval_samples_per_second": 1.668,
|
| 412 |
+
"eval_steps_per_second": 0.21,
|
| 413 |
+
"step": 270
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 18.67,
|
| 417 |
+
"learning_rate": 7.555555555555556e-05,
|
| 418 |
+
"loss": 0.0121,
|
| 419 |
+
"step": 280
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 18.67,
|
| 423 |
+
"eval_accuracy": 0.7514285714285714,
|
| 424 |
+
"eval_loss": 1.1773394346237183,
|
| 425 |
+
"eval_runtime": 220.9849,
|
| 426 |
+
"eval_samples_per_second": 1.584,
|
| 427 |
+
"eval_steps_per_second": 0.199,
|
| 428 |
+
"step": 280
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 19.33,
|
| 432 |
+
"learning_rate": 7.111111111111112e-05,
|
| 433 |
+
"loss": 0.0114,
|
| 434 |
+
"step": 290
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 19.33,
|
| 438 |
+
"eval_accuracy": 0.7514285714285714,
|
| 439 |
+
"eval_loss": 1.1831494569778442,
|
| 440 |
+
"eval_runtime": 216.1426,
|
| 441 |
+
"eval_samples_per_second": 1.619,
|
| 442 |
+
"eval_steps_per_second": 0.204,
|
| 443 |
+
"step": 290
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 20.0,
|
| 447 |
+
"learning_rate": 6.666666666666667e-05,
|
| 448 |
+
"loss": 0.0111,
|
| 449 |
+
"step": 300
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 20.0,
|
| 453 |
+
"eval_accuracy": 0.7514285714285714,
|
| 454 |
+
"eval_loss": 1.1883198022842407,
|
| 455 |
+
"eval_runtime": 210.1269,
|
| 456 |
+
"eval_samples_per_second": 1.666,
|
| 457 |
+
"eval_steps_per_second": 0.209,
|
| 458 |
+
"step": 300
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 20.67,
|
| 462 |
+
"learning_rate": 6.222222222222222e-05,
|
| 463 |
+
"loss": 0.011,
|
| 464 |
+
"step": 310
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 20.67,
|
| 468 |
+
"eval_accuracy": 0.7514285714285714,
|
| 469 |
+
"eval_loss": 1.1936721801757812,
|
| 470 |
+
"eval_runtime": 210.5377,
|
| 471 |
+
"eval_samples_per_second": 1.662,
|
| 472 |
+
"eval_steps_per_second": 0.209,
|
| 473 |
+
"step": 310
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 21.33,
|
| 477 |
+
"learning_rate": 5.7777777777777776e-05,
|
| 478 |
+
"loss": 0.0103,
|
| 479 |
+
"step": 320
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 21.33,
|
| 483 |
+
"eval_accuracy": 0.7514285714285714,
|
| 484 |
+
"eval_loss": 1.1992676258087158,
|
| 485 |
+
"eval_runtime": 214.4415,
|
| 486 |
+
"eval_samples_per_second": 1.632,
|
| 487 |
+
"eval_steps_per_second": 0.205,
|
| 488 |
+
"step": 320
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 22.0,
|
| 492 |
+
"learning_rate": 5.333333333333333e-05,
|
| 493 |
+
"loss": 0.0103,
|
| 494 |
+
"step": 330
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 22.0,
|
| 498 |
+
"eval_accuracy": 0.7514285714285714,
|
| 499 |
+
"eval_loss": 1.2045563459396362,
|
| 500 |
+
"eval_runtime": 211.851,
|
| 501 |
+
"eval_samples_per_second": 1.652,
|
| 502 |
+
"eval_steps_per_second": 0.208,
|
| 503 |
+
"step": 330
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 22.67,
|
| 507 |
+
"learning_rate": 4.888888888888889e-05,
|
| 508 |
+
"loss": 0.0103,
|
| 509 |
+
"step": 340
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 22.67,
|
| 513 |
+
"eval_accuracy": 0.7514285714285714,
|
| 514 |
+
"eval_loss": 1.208925724029541,
|
| 515 |
+
"eval_runtime": 209.6112,
|
| 516 |
+
"eval_samples_per_second": 1.67,
|
| 517 |
+
"eval_steps_per_second": 0.21,
|
| 518 |
+
"step": 340
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 23.33,
|
| 522 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 523 |
+
"loss": 0.0096,
|
| 524 |
+
"step": 350
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 23.33,
|
| 528 |
+
"eval_accuracy": 0.7514285714285714,
|
| 529 |
+
"eval_loss": 1.2132576704025269,
|
| 530 |
+
"eval_runtime": 219.2216,
|
| 531 |
+
"eval_samples_per_second": 1.597,
|
| 532 |
+
"eval_steps_per_second": 0.201,
|
| 533 |
+
"step": 350
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 24.0,
|
| 537 |
+
"learning_rate": 4e-05,
|
| 538 |
+
"loss": 0.0095,
|
| 539 |
+
"step": 360
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 24.0,
|
| 543 |
+
"eval_accuracy": 0.7514285714285714,
|
| 544 |
+
"eval_loss": 1.2171136140823364,
|
| 545 |
+
"eval_runtime": 208.9362,
|
| 546 |
+
"eval_samples_per_second": 1.675,
|
| 547 |
+
"eval_steps_per_second": 0.211,
|
| 548 |
+
"step": 360
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 24.67,
|
| 552 |
+
"learning_rate": 3.555555555555556e-05,
|
| 553 |
+
"loss": 0.0096,
|
| 554 |
+
"step": 370
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 24.67,
|
| 558 |
+
"eval_accuracy": 0.7514285714285714,
|
| 559 |
+
"eval_loss": 1.2204023599624634,
|
| 560 |
+
"eval_runtime": 209.9857,
|
| 561 |
+
"eval_samples_per_second": 1.667,
|
| 562 |
+
"eval_steps_per_second": 0.21,
|
| 563 |
+
"step": 370
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 25.33,
|
| 567 |
+
"learning_rate": 3.111111111111111e-05,
|
| 568 |
+
"loss": 0.0093,
|
| 569 |
+
"step": 380
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 25.33,
|
| 573 |
+
"eval_accuracy": 0.7485714285714286,
|
| 574 |
+
"eval_loss": 1.2234961986541748,
|
| 575 |
+
"eval_runtime": 216.2311,
|
| 576 |
+
"eval_samples_per_second": 1.619,
|
| 577 |
+
"eval_steps_per_second": 0.203,
|
| 578 |
+
"step": 380
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 26.0,
|
| 582 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 583 |
+
"loss": 0.0091,
|
| 584 |
+
"step": 390
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 26.0,
|
| 588 |
+
"eval_accuracy": 0.7485714285714286,
|
| 589 |
+
"eval_loss": 1.2261521816253662,
|
| 590 |
+
"eval_runtime": 210.1553,
|
| 591 |
+
"eval_samples_per_second": 1.665,
|
| 592 |
+
"eval_steps_per_second": 0.209,
|
| 593 |
+
"step": 390
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 26.67,
|
| 597 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 598 |
+
"loss": 0.0092,
|
| 599 |
+
"step": 400
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 26.67,
|
| 603 |
+
"eval_accuracy": 0.7514285714285714,
|
| 604 |
+
"eval_loss": 1.227960228919983,
|
| 605 |
+
"eval_runtime": 211.3289,
|
| 606 |
+
"eval_samples_per_second": 1.656,
|
| 607 |
+
"eval_steps_per_second": 0.208,
|
| 608 |
+
"step": 400
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 27.33,
|
| 612 |
+
"learning_rate": 1.777777777777778e-05,
|
| 613 |
+
"loss": 0.0089,
|
| 614 |
+
"step": 410
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 27.33,
|
| 618 |
+
"eval_accuracy": 0.7514285714285714,
|
| 619 |
+
"eval_loss": 1.2296239137649536,
|
| 620 |
+
"eval_runtime": 213.6977,
|
| 621 |
+
"eval_samples_per_second": 1.638,
|
| 622 |
+
"eval_steps_per_second": 0.206,
|
| 623 |
+
"step": 410
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 28.0,
|
| 627 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 628 |
+
"loss": 0.0092,
|
| 629 |
+
"step": 420
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 28.0,
|
| 633 |
+
"eval_accuracy": 0.7514285714285714,
|
| 634 |
+
"eval_loss": 1.2309640645980835,
|
| 635 |
+
"eval_runtime": 210.9587,
|
| 636 |
+
"eval_samples_per_second": 1.659,
|
| 637 |
+
"eval_steps_per_second": 0.209,
|
| 638 |
+
"step": 420
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 28.67,
|
| 642 |
+
"learning_rate": 8.88888888888889e-06,
|
| 643 |
+
"loss": 0.0089,
|
| 644 |
+
"step": 430
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 28.67,
|
| 648 |
+
"eval_accuracy": 0.7485714285714286,
|
| 649 |
+
"eval_loss": 1.2319449186325073,
|
| 650 |
+
"eval_runtime": 210.158,
|
| 651 |
+
"eval_samples_per_second": 1.665,
|
| 652 |
+
"eval_steps_per_second": 0.209,
|
| 653 |
+
"step": 430
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 29.33,
|
| 657 |
+
"learning_rate": 4.444444444444445e-06,
|
| 658 |
+
"loss": 0.0089,
|
| 659 |
+
"step": 440
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 29.33,
|
| 663 |
+
"eval_accuracy": 0.7485714285714286,
|
| 664 |
+
"eval_loss": 1.23252272605896,
|
| 665 |
+
"eval_runtime": 216.1534,
|
| 666 |
+
"eval_samples_per_second": 1.619,
|
| 667 |
+
"eval_steps_per_second": 0.204,
|
| 668 |
+
"step": 440
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 30.0,
|
| 672 |
+
"learning_rate": 0.0,
|
| 673 |
+
"loss": 0.0088,
|
| 674 |
+
"step": 450
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 30.0,
|
| 678 |
+
"eval_accuracy": 0.7485714285714286,
|
| 679 |
+
"eval_loss": 1.2327271699905396,
|
| 680 |
+
"eval_runtime": 216.0649,
|
| 681 |
+
"eval_samples_per_second": 1.62,
|
| 682 |
+
"eval_steps_per_second": 0.204,
|
| 683 |
+
"step": 450
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 30.0,
|
| 687 |
+
"step": 450,
|
| 688 |
+
"total_flos": 9.8805828893184e+17,
|
| 689 |
+
"train_loss": 0.09906705205639203,
|
| 690 |
+
"train_runtime": 36217.9644,
|
| 691 |
+
"train_samples_per_second": 0.352,
|
| 692 |
+
"train_steps_per_second": 0.012
|
| 693 |
+
}
|
| 694 |
+
],
|
| 695 |
+
"max_steps": 450,
|
| 696 |
+
"num_train_epochs": 30,
|
| 697 |
+
"total_flos": 9.8805828893184e+17,
|
| 698 |
+
"trial_name": null,
|
| 699 |
+
"trial_params": null
|
| 700 |
+
}
|