Upload trainer_state.json
Browse files- trainer_state.json +96 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5530,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9,
|
| 13 |
+
"learning_rate": 4.547920433996384e-05,
|
| 14 |
+
"loss": 2.8592,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.81,
|
| 19 |
+
"learning_rate": 4.095840867992767e-05,
|
| 20 |
+
"loss": 1.9624,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 2.71,
|
| 25 |
+
"learning_rate": 3.6437613019891505e-05,
|
| 26 |
+
"loss": 1.5727,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 3.62,
|
| 31 |
+
"learning_rate": 3.191681735985534e-05,
|
| 32 |
+
"loss": 1.3126,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 4.52,
|
| 37 |
+
"learning_rate": 2.7396021699819167e-05,
|
| 38 |
+
"loss": 1.1791,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 5.42,
|
| 43 |
+
"learning_rate": 2.2875226039783005e-05,
|
| 44 |
+
"loss": 0.9823,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 6.33,
|
| 49 |
+
"learning_rate": 1.8354430379746836e-05,
|
| 50 |
+
"loss": 0.8947,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 7.23,
|
| 55 |
+
"learning_rate": 1.383363471971067e-05,
|
| 56 |
+
"loss": 0.7946,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 8.14,
|
| 61 |
+
"learning_rate": 9.312839059674505e-06,
|
| 62 |
+
"loss": 0.7585,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 9.04,
|
| 67 |
+
"learning_rate": 4.792043399638336e-06,
|
| 68 |
+
"loss": 0.6896,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 9.95,
|
| 73 |
+
"learning_rate": 2.7124773960217e-07,
|
| 74 |
+
"loss": 0.6409,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 10.0,
|
| 79 |
+
"step": 5530,
|
| 80 |
+
"total_flos": 2999368568832000.0,
|
| 81 |
+
"train_loss": 1.237921866542823,
|
| 82 |
+
"train_runtime": 1131.6305,
|
| 83 |
+
"train_samples_per_second": 19.538,
|
| 84 |
+
"train_steps_per_second": 4.887
|
| 85 |
+
}
|
| 86 |
+
],
|
| 87 |
+
"logging_steps": 500,
|
| 88 |
+
"max_steps": 5530,
|
| 89 |
+
"num_input_tokens_seen": 0,
|
| 90 |
+
"num_train_epochs": 10,
|
| 91 |
+
"save_steps": 500,
|
| 92 |
+
"total_flos": 2999368568832000.0,
|
| 93 |
+
"train_batch_size": 4,
|
| 94 |
+
"trial_name": null,
|
| 95 |
+
"trial_params": null
|
| 96 |
+
}
|