Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +118 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,118 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 15.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 12585,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.1918951132300357,
|
| 14 |
+
"grad_norm": 0.8493185043334961,
|
| 15 |
+
"learning_rate": 0.00019872148733408694,
|
| 16 |
+
"loss": 0.5421,
|
| 17 |
+
"step": 1000
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 2.3837902264600714,
|
| 21 |
+
"grad_norm": 0.4860774874687195,
|
| 22 |
+
"learning_rate": 0.00019141307038722188,
|
| 23 |
+
"loss": 0.2042,
|
| 24 |
+
"step": 2000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 3.575685339690107,
|
| 28 |
+
"grad_norm": 1.4351465702056885,
|
| 29 |
+
"learning_rate": 0.00017808333307050135,
|
| 30 |
+
"loss": 0.1613,
|
| 31 |
+
"step": 3000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 4.767580452920143,
|
| 35 |
+
"grad_norm": 0.7845919728279114,
|
| 36 |
+
"learning_rate": 0.00015961029663941979,
|
| 37 |
+
"loss": 0.1339,
|
| 38 |
+
"step": 4000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 5.959475566150179,
|
| 42 |
+
"grad_norm": 0.16761158406734467,
|
| 43 |
+
"learning_rate": 0.00013721076816636165,
|
| 44 |
+
"loss": 0.1194,
|
| 45 |
+
"step": 5000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 7.151370679380214,
|
| 49 |
+
"grad_norm": 0.18906846642494202,
|
| 50 |
+
"learning_rate": 0.00011236019023846099,
|
| 51 |
+
"loss": 0.1014,
|
| 52 |
+
"step": 6000
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 8.34326579261025,
|
| 56 |
+
"grad_norm": 0.3582542836666107,
|
| 57 |
+
"learning_rate": 8.669545449983509e-05,
|
| 58 |
+
"loss": 0.0894,
|
| 59 |
+
"step": 7000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 9.535160905840286,
|
| 63 |
+
"grad_norm": 1.641005277633667,
|
| 64 |
+
"learning_rate": 6.190708064788978e-05,
|
| 65 |
+
"loss": 0.0815,
|
| 66 |
+
"step": 8000
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 10.727056019070321,
|
| 70 |
+
"grad_norm": 0.1235262081027031,
|
| 71 |
+
"learning_rate": 3.962786298411891e-05,
|
| 72 |
+
"loss": 0.0732,
|
| 73 |
+
"step": 9000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 11.918951132300357,
|
| 77 |
+
"grad_norm": 0.12703625857830048,
|
| 78 |
+
"learning_rate": 2.132531929974263e-05,
|
| 79 |
+
"loss": 0.0674,
|
| 80 |
+
"step": 10000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 13.110846245530393,
|
| 84 |
+
"grad_norm": 0.35737335681915283,
|
| 85 |
+
"learning_rate": 8.20502641916997e-06,
|
| 86 |
+
"loss": 0.0645,
|
| 87 |
+
"step": 11000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 14.302741358760429,
|
| 91 |
+
"grad_norm": 0.10338258743286133,
|
| 92 |
+
"learning_rate": 1.1312096270961526e-06,
|
| 93 |
+
"loss": 0.0623,
|
| 94 |
+
"step": 12000
|
| 95 |
+
}
|
| 96 |
+
],
|
| 97 |
+
"logging_steps": 1000,
|
| 98 |
+
"max_steps": 12585,
|
| 99 |
+
"num_input_tokens_seen": 0,
|
| 100 |
+
"num_train_epochs": 15,
|
| 101 |
+
"save_steps": 10000,
|
| 102 |
+
"stateful_callbacks": {
|
| 103 |
+
"TrainerControl": {
|
| 104 |
+
"args": {
|
| 105 |
+
"should_epoch_stop": false,
|
| 106 |
+
"should_evaluate": false,
|
| 107 |
+
"should_log": false,
|
| 108 |
+
"should_save": true,
|
| 109 |
+
"should_training_stop": true
|
| 110 |
+
},
|
| 111 |
+
"attributes": {}
|
| 112 |
+
}
|
| 113 |
+
},
|
| 114 |
+
"total_flos": 1.0572731031172128e+17,
|
| 115 |
+
"train_batch_size": 2,
|
| 116 |
+
"trial_name": null,
|
| 117 |
+
"trial_params": null
|
| 118 |
+
}
|