Add training logs and README
Browse files
checkpoint-1176/trainer_state.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1176,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.850792300329682,
|
| 14 |
+
"grad_norm": 0.338798850774765,
|
| 15 |
+
"learning_rate": 6.635742513724948e-06,
|
| 16 |
+
"loss": 0.7492,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.701052855471658,
|
| 21 |
+
"grad_norm": 0.5019412636756897,
|
| 22 |
+
"learning_rate": 6.068653842993466e-07,
|
| 23 |
+
"loss": 0.6907,
|
| 24 |
+
"step": 1000
|
| 25 |
+
}
|
| 26 |
+
],
|
| 27 |
+
"logging_steps": 500,
|
| 28 |
+
"max_steps": 1176,
|
| 29 |
+
"num_input_tokens_seen": 0,
|
| 30 |
+
"num_train_epochs": 2,
|
| 31 |
+
"save_steps": 250,
|
| 32 |
+
"stateful_callbacks": {
|
| 33 |
+
"TrainerControl": {
|
| 34 |
+
"args": {
|
| 35 |
+
"should_epoch_stop": false,
|
| 36 |
+
"should_evaluate": false,
|
| 37 |
+
"should_log": false,
|
| 38 |
+
"should_save": true,
|
| 39 |
+
"should_training_stop": true
|
| 40 |
+
},
|
| 41 |
+
"attributes": {}
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"total_flos": 2.4116995620217405e+18,
|
| 45 |
+
"train_batch_size": 1,
|
| 46 |
+
"trial_name": null,
|
| 47 |
+
"trial_params": null
|
| 48 |
+
}
|