Upload trainer_state.json
Browse files- trainer_state.json +85 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.24,
|
| 5 |
+
"global_step": 30,
|
| 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": 4.8400000000000004e-05,
|
| 13 |
+
"loss": 2.9182,
|
| 14 |
+
"step": 5
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"learning_rate": 4.64e-05,
|
| 19 |
+
"loss": 2.4304,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.08,
|
| 24 |
+
"eval_loss": 2.1126761436462402,
|
| 25 |
+
"eval_rouge2_fmeasure": 0.2505,
|
| 26 |
+
"eval_rouge2_precision": 0.2538,
|
| 27 |
+
"eval_rouge2_recall": 0.2509,
|
| 28 |
+
"eval_runtime": 13283.7379,
|
| 29 |
+
"eval_samples_per_second": 0.023,
|
| 30 |
+
"eval_steps_per_second": 0.011,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.12,
|
| 35 |
+
"learning_rate": 4.44e-05,
|
| 36 |
+
"loss": 2.2949,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16,
|
| 41 |
+
"learning_rate": 4.24e-05,
|
| 42 |
+
"loss": 2.145,
|
| 43 |
+
"step": 20
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.16,
|
| 47 |
+
"eval_loss": 1.9695031642913818,
|
| 48 |
+
"eval_rouge2_fmeasure": 0.2588,
|
| 49 |
+
"eval_rouge2_precision": 0.2769,
|
| 50 |
+
"eval_rouge2_recall": 0.2515,
|
| 51 |
+
"eval_runtime": 13236.2799,
|
| 52 |
+
"eval_samples_per_second": 0.023,
|
| 53 |
+
"eval_steps_per_second": 0.011,
|
| 54 |
+
"step": 20
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.2,
|
| 58 |
+
"learning_rate": 4.0400000000000006e-05,
|
| 59 |
+
"loss": 2.1084,
|
| 60 |
+
"step": 25
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.24,
|
| 64 |
+
"learning_rate": 3.8400000000000005e-05,
|
| 65 |
+
"loss": 2.1794,
|
| 66 |
+
"step": 30
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.24,
|
| 70 |
+
"eval_loss": 1.8257582187652588,
|
| 71 |
+
"eval_rouge2_fmeasure": 0.2789,
|
| 72 |
+
"eval_rouge2_precision": 0.2837,
|
| 73 |
+
"eval_rouge2_recall": 0.2784,
|
| 74 |
+
"eval_runtime": 13267.4385,
|
| 75 |
+
"eval_samples_per_second": 0.023,
|
| 76 |
+
"eval_steps_per_second": 0.011,
|
| 77 |
+
"step": 30
|
| 78 |
+
}
|
| 79 |
+
],
|
| 80 |
+
"max_steps": 125,
|
| 81 |
+
"num_train_epochs": 1,
|
| 82 |
+
"total_flos": 1296098320711680.0,
|
| 83 |
+
"trial_name": null,
|
| 84 |
+
"trial_params": null
|
| 85 |
+
}
|