Upload trainer_state.json
Browse files- trainer_state.json +72 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 3608,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"learning_rate": 4.166923313828149e-06,
|
| 13 |
+
"loss": 2.6072,
|
| 14 |
+
"step": 902
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 2.3964672088623047,
|
| 19 |
+
"eval_runtime": 18.2149,
|
| 20 |
+
"eval_samples_per_second": 54.9,
|
| 21 |
+
"eval_steps_per_second": 0.604,
|
| 22 |
+
"step": 902
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.0,
|
| 26 |
+
"learning_rate": 2.777948875885433e-06,
|
| 27 |
+
"loss": 2.522,
|
| 28 |
+
"step": 1804
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"eval_loss": 2.383113384246826,
|
| 33 |
+
"eval_runtime": 1.6537,
|
| 34 |
+
"eval_samples_per_second": 604.697,
|
| 35 |
+
"eval_steps_per_second": 6.652,
|
| 36 |
+
"step": 1804
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.0,
|
| 40 |
+
"learning_rate": 1.3889744379427164e-06,
|
| 41 |
+
"loss": 2.485,
|
| 42 |
+
"step": 2706
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.0,
|
| 46 |
+
"eval_loss": 2.3780102729797363,
|
| 47 |
+
"eval_runtime": 1.6082,
|
| 48 |
+
"eval_samples_per_second": 621.809,
|
| 49 |
+
"eval_steps_per_second": 6.84,
|
| 50 |
+
"step": 2706
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"learning_rate": 0.0,
|
| 55 |
+
"loss": 2.4641,
|
| 56 |
+
"step": 3608
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.0,
|
| 60 |
+
"eval_loss": 2.3768954277038574,
|
| 61 |
+
"eval_runtime": 1.6566,
|
| 62 |
+
"eval_samples_per_second": 603.647,
|
| 63 |
+
"eval_steps_per_second": 6.64,
|
| 64 |
+
"step": 3608
|
| 65 |
+
}
|
| 66 |
+
],
|
| 67 |
+
"max_steps": 3608,
|
| 68 |
+
"num_train_epochs": 4,
|
| 69 |
+
"total_flos": 1.1301011030016e+16,
|
| 70 |
+
"trial_name": null,
|
| 71 |
+
"trial_params": null
|
| 72 |
+
}
|