Upload trainer_state.json
Browse files- trainer_state.json +131 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,131 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.4,
|
| 5 |
+
"global_step": 50,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"learning_rate": 4.8e-05,
|
| 13 |
+
"loss": 2.3721,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"learning_rate": 4.600000000000001e-05,
|
| 19 |
+
"loss": 1.8562,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.08,
|
| 24 |
+
"eval_loss": 1.995707631111145,
|
| 25 |
+
"eval_rouge2_fmeasure": 0.2129,
|
| 26 |
+
"eval_rouge2_precision": 0.2635,
|
| 27 |
+
"eval_rouge2_recall": 0.2168,
|
| 28 |
+
"eval_runtime": 3900.1086,
|
| 29 |
+
"eval_samples_per_second": 0.077,
|
| 30 |
+
"eval_steps_per_second": 0.038,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.12,
|
| 35 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 36 |
+
"loss": 2.1063,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16,
|
| 41 |
+
"learning_rate": 4.2e-05,
|
| 42 |
+
"loss": 1.8923,
|
| 43 |
+
"step": 20
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.16,
|
| 47 |
+
"eval_loss": 1.7682132720947266,
|
| 48 |
+
"eval_rouge2_fmeasure": 0.2466,
|
| 49 |
+
"eval_rouge2_precision": 0.2599,
|
| 50 |
+
"eval_rouge2_recall": 0.3043,
|
| 51 |
+
"eval_runtime": 7803.2087,
|
| 52 |
+
"eval_samples_per_second": 0.038,
|
| 53 |
+
"eval_steps_per_second": 0.019,
|
| 54 |
+
"step": 20
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.2,
|
| 58 |
+
"learning_rate": 4e-05,
|
| 59 |
+
"loss": 1.7652,
|
| 60 |
+
"step": 25
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.24,
|
| 64 |
+
"learning_rate": 3.8e-05,
|
| 65 |
+
"loss": 1.909,
|
| 66 |
+
"step": 30
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.24,
|
| 70 |
+
"eval_loss": 1.7092610597610474,
|
| 71 |
+
"eval_rouge2_fmeasure": 0.2531,
|
| 72 |
+
"eval_rouge2_precision": 0.2638,
|
| 73 |
+
"eval_rouge2_recall": 0.3093,
|
| 74 |
+
"eval_runtime": 7685.9909,
|
| 75 |
+
"eval_samples_per_second": 0.039,
|
| 76 |
+
"eval_steps_per_second": 0.02,
|
| 77 |
+
"step": 30
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.28,
|
| 81 |
+
"learning_rate": 3.6e-05,
|
| 82 |
+
"loss": 1.686,
|
| 83 |
+
"step": 35
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.32,
|
| 87 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 88 |
+
"loss": 1.6912,
|
| 89 |
+
"step": 40
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.32,
|
| 93 |
+
"eval_loss": 1.66864812374115,
|
| 94 |
+
"eval_rouge2_fmeasure": 0.2679,
|
| 95 |
+
"eval_rouge2_precision": 0.3245,
|
| 96 |
+
"eval_rouge2_recall": 0.2789,
|
| 97 |
+
"eval_runtime": 3975.5587,
|
| 98 |
+
"eval_samples_per_second": 0.075,
|
| 99 |
+
"eval_steps_per_second": 0.038,
|
| 100 |
+
"step": 40
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.36,
|
| 104 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 105 |
+
"loss": 1.7482,
|
| 106 |
+
"step": 45
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.4,
|
| 110 |
+
"learning_rate": 3e-05,
|
| 111 |
+
"loss": 1.7148,
|
| 112 |
+
"step": 50
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.4,
|
| 116 |
+
"eval_loss": 1.6291550397872925,
|
| 117 |
+
"eval_rouge2_fmeasure": 0.267,
|
| 118 |
+
"eval_rouge2_precision": 0.3113,
|
| 119 |
+
"eval_rouge2_recall": 0.2944,
|
| 120 |
+
"eval_runtime": 5016.1567,
|
| 121 |
+
"eval_samples_per_second": 0.06,
|
| 122 |
+
"eval_steps_per_second": 0.03,
|
| 123 |
+
"step": 50
|
| 124 |
+
}
|
| 125 |
+
],
|
| 126 |
+
"max_steps": 125,
|
| 127 |
+
"num_train_epochs": 1,
|
| 128 |
+
"total_flos": 2160163867852800.0,
|
| 129 |
+
"trial_name": null,
|
| 130 |
+
"trial_params": null
|
| 131 |
+
}
|