Upload trainer_state.json
Browse files- trainer_state.json +73 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.5944311618804932,
|
| 3 |
+
"best_model_checkpoint": "/nfs/paraphrasing/output/bert-finetuned-para/checkpoint-1500",
|
| 4 |
+
"epoch": 0.008791827117511562,
|
| 5 |
+
"global_step": 1500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 1.9998827756384335e-05,
|
| 13 |
+
"loss": 2.0282,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.0,
|
| 18 |
+
"eval_gen_len": 13.1456,
|
| 19 |
+
"eval_loss": 1.6860297918319702,
|
| 20 |
+
"eval_rouge1": 55.3995,
|
| 21 |
+
"eval_rouge2": 32.0094,
|
| 22 |
+
"eval_rougeL": 53.1553,
|
| 23 |
+
"eval_rougeLsum": 53.1706,
|
| 24 |
+
"eval_runtime": 50298.9522,
|
| 25 |
+
"eval_samples_per_second": 27.136,
|
| 26 |
+
"eval_steps_per_second": 0.424,
|
| 27 |
+
"step": 500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.01,
|
| 31 |
+
"learning_rate": 1.9997655512768664e-05,
|
| 32 |
+
"loss": 1.8116,
|
| 33 |
+
"step": 1000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.01,
|
| 37 |
+
"eval_gen_len": 13.1187,
|
| 38 |
+
"eval_loss": 1.6253910064697266,
|
| 39 |
+
"eval_rouge1": 56.0224,
|
| 40 |
+
"eval_rouge2": 32.6912,
|
| 41 |
+
"eval_rougeL": 53.7965,
|
| 42 |
+
"eval_rougeLsum": 53.8115,
|
| 43 |
+
"eval_runtime": 111894.1395,
|
| 44 |
+
"eval_samples_per_second": 12.198,
|
| 45 |
+
"eval_steps_per_second": 0.191,
|
| 46 |
+
"step": 1000
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.01,
|
| 50 |
+
"learning_rate": 1.9996483269152997e-05,
|
| 51 |
+
"loss": 1.7615,
|
| 52 |
+
"step": 1500
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.01,
|
| 56 |
+
"eval_gen_len": 13.0796,
|
| 57 |
+
"eval_loss": 1.5944311618804932,
|
| 58 |
+
"eval_rouge1": 56.3476,
|
| 59 |
+
"eval_rouge2": 33.0715,
|
| 60 |
+
"eval_rougeL": 54.1323,
|
| 61 |
+
"eval_rougeLsum": 54.1475,
|
| 62 |
+
"eval_runtime": 102094.2868,
|
| 63 |
+
"eval_samples_per_second": 13.369,
|
| 64 |
+
"eval_steps_per_second": 0.209,
|
| 65 |
+
"step": 1500
|
| 66 |
+
}
|
| 67 |
+
],
|
| 68 |
+
"max_steps": 8530650,
|
| 69 |
+
"num_train_epochs": 50,
|
| 70 |
+
"total_flos": 4699507207962624.0,
|
| 71 |
+
"trial_name": null,
|
| 72 |
+
"trial_params": null
|
| 73 |
+
}
|