Upload checkpoint-308/trainer_state.json
Browse files
checkpoint-308/trainer_state.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 308,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.16260162601626016,
|
| 14 |
+
"grad_norm": 0.37675046920776367,
|
| 15 |
+
"learning_rate": 0.0001681818181818182,
|
| 16 |
+
"loss": 1.7515,
|
| 17 |
+
"mean_token_accuracy": 0.6049755607545376,
|
| 18 |
+
"num_tokens": 204800.0,
|
| 19 |
+
"step": 50
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.3252032520325203,
|
| 23 |
+
"grad_norm": 0.3089984357357025,
|
| 24 |
+
"learning_rate": 0.00013571428571428572,
|
| 25 |
+
"loss": 1.3323,
|
| 26 |
+
"mean_token_accuracy": 0.6925122183561325,
|
| 27 |
+
"num_tokens": 409600.0,
|
| 28 |
+
"step": 100
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.4878048780487805,
|
| 32 |
+
"grad_norm": 0.3363010883331299,
|
| 33 |
+
"learning_rate": 0.00010324675324675325,
|
| 34 |
+
"loss": 1.2756,
|
| 35 |
+
"mean_token_accuracy": 0.7040957960486413,
|
| 36 |
+
"num_tokens": 614400.0,
|
| 37 |
+
"step": 150
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.6504065040650406,
|
| 41 |
+
"grad_norm": 0.335611492395401,
|
| 42 |
+
"learning_rate": 7.077922077922077e-05,
|
| 43 |
+
"loss": 1.2833,
|
| 44 |
+
"mean_token_accuracy": 0.7033577680587768,
|
| 45 |
+
"num_tokens": 819200.0,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.8130081300813008,
|
| 50 |
+
"grad_norm": 0.3524633049964905,
|
| 51 |
+
"learning_rate": 3.831168831168831e-05,
|
| 52 |
+
"loss": 1.2286,
|
| 53 |
+
"mean_token_accuracy": 0.7145161288976669,
|
| 54 |
+
"num_tokens": 1024000.0,
|
| 55 |
+
"step": 250
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.975609756097561,
|
| 59 |
+
"grad_norm": 0.34292736649513245,
|
| 60 |
+
"learning_rate": 5.844155844155844e-06,
|
| 61 |
+
"loss": 1.2549,
|
| 62 |
+
"mean_token_accuracy": 0.7073900297284126,
|
| 63 |
+
"num_tokens": 1228800.0,
|
| 64 |
+
"step": 300
|
| 65 |
+
}
|
| 66 |
+
],
|
| 67 |
+
"logging_steps": 50,
|
| 68 |
+
"max_steps": 308,
|
| 69 |
+
"num_input_tokens_seen": 0,
|
| 70 |
+
"num_train_epochs": 1,
|
| 71 |
+
"save_steps": 200,
|
| 72 |
+
"stateful_callbacks": {
|
| 73 |
+
"TrainerControl": {
|
| 74 |
+
"args": {
|
| 75 |
+
"should_epoch_stop": false,
|
| 76 |
+
"should_evaluate": false,
|
| 77 |
+
"should_log": false,
|
| 78 |
+
"should_save": true,
|
| 79 |
+
"should_training_stop": true
|
| 80 |
+
},
|
| 81 |
+
"attributes": {}
|
| 82 |
+
}
|
| 83 |
+
},
|
| 84 |
+
"total_flos": 2.09971568443392e+16,
|
| 85 |
+
"train_batch_size": 1,
|
| 86 |
+
"trial_name": null,
|
| 87 |
+
"trial_params": null
|
| 88 |
+
}
|