Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +51 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1857,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.8077544426494345,
|
| 13 |
+
"grad_norm": 2.0110491628996834,
|
| 14 |
+
"learning_rate": 3.653742595584276e-05,
|
| 15 |
+
"loss": 2.576,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.615508885298869,
|
| 20 |
+
"grad_norm": 1.922235177894796,
|
| 21 |
+
"learning_rate": 2.3074851911685516e-05,
|
| 22 |
+
"loss": 1.9215,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.4232633279483036,
|
| 27 |
+
"grad_norm": 2.3049251003554425,
|
| 28 |
+
"learning_rate": 9.612277867528271e-06,
|
| 29 |
+
"loss": 1.4342,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 3.0,
|
| 34 |
+
"step": 1857,
|
| 35 |
+
"total_flos": 5.480111913004171e+18,
|
| 36 |
+
"train_loss": 1.8135224785545403,
|
| 37 |
+
"train_runtime": 5570.2565,
|
| 38 |
+
"train_samples_per_second": 21.323,
|
| 39 |
+
"train_steps_per_second": 0.333
|
| 40 |
+
}
|
| 41 |
+
],
|
| 42 |
+
"logging_steps": 500,
|
| 43 |
+
"max_steps": 1857,
|
| 44 |
+
"num_input_tokens_seen": 0,
|
| 45 |
+
"num_train_epochs": 3,
|
| 46 |
+
"save_steps": 500,
|
| 47 |
+
"total_flos": 5.480111913004171e+18,
|
| 48 |
+
"train_batch_size": 8,
|
| 49 |
+
"trial_name": null,
|
| 50 |
+
"trial_params": null
|
| 51 |
+
}
|