Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +64 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 30,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 3.3333333333333335,
|
| 14 |
+
"grad_norm": 1.8390942811965942,
|
| 15 |
+
"learning_rate": 8.535533905932738e-05,
|
| 16 |
+
"loss": 0.5065,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 6.666666666666667,
|
| 21 |
+
"grad_norm": 0.7160496115684509,
|
| 22 |
+
"learning_rate": 3.3486046902241664e-05,
|
| 23 |
+
"loss": 0.2128,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 10.0,
|
| 28 |
+
"grad_norm": 0.446665495634079,
|
| 29 |
+
"learning_rate": 3.143895053378698e-07,
|
| 30 |
+
"loss": 0.1702,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 10.0,
|
| 35 |
+
"step": 30,
|
| 36 |
+
"total_flos": 0.0,
|
| 37 |
+
"train_loss": 0.29649561246236167,
|
| 38 |
+
"train_runtime": 65.8291,
|
| 39 |
+
"train_samples_per_second": 47.547,
|
| 40 |
+
"train_steps_per_second": 0.456
|
| 41 |
+
}
|
| 42 |
+
],
|
| 43 |
+
"logging_steps": 10,
|
| 44 |
+
"max_steps": 30,
|
| 45 |
+
"num_input_tokens_seen": 0,
|
| 46 |
+
"num_train_epochs": 10,
|
| 47 |
+
"save_steps": 10000,
|
| 48 |
+
"stateful_callbacks": {
|
| 49 |
+
"TrainerControl": {
|
| 50 |
+
"args": {
|
| 51 |
+
"should_epoch_stop": false,
|
| 52 |
+
"should_evaluate": false,
|
| 53 |
+
"should_log": false,
|
| 54 |
+
"should_save": true,
|
| 55 |
+
"should_training_stop": true
|
| 56 |
+
},
|
| 57 |
+
"attributes": {}
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
"total_flos": 0.0,
|
| 61 |
+
"train_batch_size": 107,
|
| 62 |
+
"trial_name": null,
|
| 63 |
+
"trial_params": null
|
| 64 |
+
}
|