Upload trainer_state.json
Browse files- trainer_state.json +100 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,100 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 18850,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"learning_rate": 4.4446191337159036e-05,
|
| 13 |
+
"loss": 3.7069,
|
| 14 |
+
"step": 2827
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 3.1980156898498535,
|
| 19 |
+
"eval_runtime": 24.6141,
|
| 20 |
+
"eval_samples_per_second": 200.048,
|
| 21 |
+
"eval_steps_per_second": 1.584,
|
| 22 |
+
"step": 2827
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.0,
|
| 26 |
+
"learning_rate": 3.333464350286927e-05,
|
| 27 |
+
"loss": 3.0719,
|
| 28 |
+
"step": 5654
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"eval_loss": 3.040849447250366,
|
| 33 |
+
"eval_runtime": 7.1854,
|
| 34 |
+
"eval_samples_per_second": 685.277,
|
| 35 |
+
"eval_steps_per_second": 5.428,
|
| 36 |
+
"step": 5654
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.0,
|
| 40 |
+
"learning_rate": 2.2223095668579518e-05,
|
| 41 |
+
"loss": 2.8585,
|
| 42 |
+
"step": 8481
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.0,
|
| 46 |
+
"eval_loss": 2.969093084335327,
|
| 47 |
+
"eval_runtime": 7.1989,
|
| 48 |
+
"eval_samples_per_second": 683.998,
|
| 49 |
+
"eval_steps_per_second": 5.418,
|
| 50 |
+
"step": 8481
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 55 |
+
"loss": 2.7462,
|
| 56 |
+
"step": 11310
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.0,
|
| 60 |
+
"eval_loss": 2.61379075050354,
|
| 61 |
+
"eval_runtime": 25.0394,
|
| 62 |
+
"eval_samples_per_second": 196.57,
|
| 63 |
+
"eval_steps_per_second": 2.077,
|
| 64 |
+
"step": 11310
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 69 |
+
"loss": 2.6518,
|
| 70 |
+
"step": 15080
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 4.0,
|
| 74 |
+
"eval_loss": 2.5971875190734863,
|
| 75 |
+
"eval_runtime": 14.767,
|
| 76 |
+
"eval_samples_per_second": 333.31,
|
| 77 |
+
"eval_steps_per_second": 3.521,
|
| 78 |
+
"step": 15080
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 5.0,
|
| 82 |
+
"learning_rate": 0.0,
|
| 83 |
+
"loss": 2.5587,
|
| 84 |
+
"step": 18850
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.0,
|
| 88 |
+
"eval_loss": 2.5887153148651123,
|
| 89 |
+
"eval_runtime": 7.6626,
|
| 90 |
+
"eval_samples_per_second": 642.341,
|
| 91 |
+
"eval_steps_per_second": 6.786,
|
| 92 |
+
"step": 18850
|
| 93 |
+
}
|
| 94 |
+
],
|
| 95 |
+
"max_steps": 18850,
|
| 96 |
+
"num_train_epochs": 5,
|
| 97 |
+
"total_flos": 6.795617845248e+16,
|
| 98 |
+
"trial_name": null,
|
| 99 |
+
"trial_params": null
|
| 100 |
+
}
|