Upload trainer_state.json
Browse files- trainer_state.json +115 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9998104265402845,
|
| 5 |
+
"global_step": 7911,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.19,
|
| 12 |
+
"learning_rate": 4.683984325622551e-05,
|
| 13 |
+
"loss": 0.7763,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.38,
|
| 18 |
+
"learning_rate": 4.367968651245102e-05,
|
| 19 |
+
"loss": 0.4148,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.57,
|
| 24 |
+
"learning_rate": 4.051952976867653e-05,
|
| 25 |
+
"loss": 0.3387,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.76,
|
| 30 |
+
"learning_rate": 3.7359373024902036e-05,
|
| 31 |
+
"loss": 0.2855,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.95,
|
| 36 |
+
"learning_rate": 3.4199216281127545e-05,
|
| 37 |
+
"loss": 0.2469,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.14,
|
| 42 |
+
"learning_rate": 3.1039059537353053e-05,
|
| 43 |
+
"loss": 0.2117,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.33,
|
| 48 |
+
"learning_rate": 2.7878902793578565e-05,
|
| 49 |
+
"loss": 0.1833,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.52,
|
| 54 |
+
"learning_rate": 2.471874604980407e-05,
|
| 55 |
+
"loss": 0.1696,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.71,
|
| 60 |
+
"learning_rate": 2.1558589306029582e-05,
|
| 61 |
+
"loss": 0.1554,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.9,
|
| 66 |
+
"learning_rate": 1.8398432562255087e-05,
|
| 67 |
+
"loss": 0.1489,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.09,
|
| 72 |
+
"learning_rate": 1.5238275818480596e-05,
|
| 73 |
+
"loss": 0.1319,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.28,
|
| 78 |
+
"learning_rate": 1.2078119074706106e-05,
|
| 79 |
+
"loss": 0.1197,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.46,
|
| 84 |
+
"learning_rate": 8.917962330931615e-06,
|
| 85 |
+
"loss": 0.1158,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.65,
|
| 90 |
+
"learning_rate": 5.757805587157123e-06,
|
| 91 |
+
"loss": 0.1105,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.84,
|
| 96 |
+
"learning_rate": 2.597648843382632e-06,
|
| 97 |
+
"loss": 0.1057,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.0,
|
| 102 |
+
"step": 7911,
|
| 103 |
+
"total_flos": 2.962265039645184e+17,
|
| 104 |
+
"train_loss": 0.22759322628271503,
|
| 105 |
+
"train_runtime": 16489.6196,
|
| 106 |
+
"train_samples_per_second": 30.708,
|
| 107 |
+
"train_steps_per_second": 0.48
|
| 108 |
+
}
|
| 109 |
+
],
|
| 110 |
+
"max_steps": 7911,
|
| 111 |
+
"num_train_epochs": 3,
|
| 112 |
+
"total_flos": 2.962265039645184e+17,
|
| 113 |
+
"trial_name": null,
|
| 114 |
+
"trial_params": null
|
| 115 |
+
}
|