Upload trainer_state.json from checkpoint
Browse files- trainer_state.json +89 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.773824980524537,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 400000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 2.596728122565567,
|
| 13 |
+
"grad_norm": 0.9470382332801819,
|
| 14 |
+
"learning_rate": 0.0002,
|
| 15 |
+
"loss": 1.7054,
|
| 16 |
+
"step": 50000
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 5.193456245131134,
|
| 20 |
+
"grad_norm": 0.6922363638877869,
|
| 21 |
+
"learning_rate": 0.0002,
|
| 22 |
+
"loss": 1.6798,
|
| 23 |
+
"step": 100000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 7.790184367696702,
|
| 27 |
+
"grad_norm": 0.6903401017189026,
|
| 28 |
+
"learning_rate": 0.0002,
|
| 29 |
+
"loss": 1.6738,
|
| 30 |
+
"step": 150000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 10.386912490262269,
|
| 34 |
+
"grad_norm": 1.2185564041137695,
|
| 35 |
+
"learning_rate": 0.0002,
|
| 36 |
+
"loss": 1.6705,
|
| 37 |
+
"step": 200000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 12.983640612827838,
|
| 41 |
+
"grad_norm": 0.7960174083709717,
|
| 42 |
+
"learning_rate": 0.0002,
|
| 43 |
+
"loss": 1.6691,
|
| 44 |
+
"step": 250000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 15.580368735393405,
|
| 48 |
+
"grad_norm": 0.7023811936378479,
|
| 49 |
+
"learning_rate": 0.0002,
|
| 50 |
+
"loss": 1.6667,
|
| 51 |
+
"step": 300000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 18.177096857958972,
|
| 55 |
+
"grad_norm": 0.6432592868804932,
|
| 56 |
+
"learning_rate": 0.0002,
|
| 57 |
+
"loss": 1.6664,
|
| 58 |
+
"step": 350000
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 20.773824980524537,
|
| 62 |
+
"grad_norm": 1.1296038627624512,
|
| 63 |
+
"learning_rate": 0.0002,
|
| 64 |
+
"loss": 1.6651,
|
| 65 |
+
"step": 400000
|
| 66 |
+
}
|
| 67 |
+
],
|
| 68 |
+
"logging_steps": 50000,
|
| 69 |
+
"max_steps": 577650,
|
| 70 |
+
"num_input_tokens_seen": 0,
|
| 71 |
+
"num_train_epochs": 30,
|
| 72 |
+
"save_steps": 50000,
|
| 73 |
+
"stateful_callbacks": {
|
| 74 |
+
"TrainerControl": {
|
| 75 |
+
"args": {
|
| 76 |
+
"should_epoch_stop": false,
|
| 77 |
+
"should_evaluate": false,
|
| 78 |
+
"should_log": false,
|
| 79 |
+
"should_save": true,
|
| 80 |
+
"should_training_stop": false
|
| 81 |
+
},
|
| 82 |
+
"attributes": {}
|
| 83 |
+
}
|
| 84 |
+
},
|
| 85 |
+
"total_flos": 4.227652833030557e+19,
|
| 86 |
+
"train_batch_size": 8,
|
| 87 |
+
"trial_name": null,
|
| 88 |
+
"trial_params": null
|
| 89 |
+
}
|