Add trainer_state.json
Browse files- trainer_state.json +104 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.0462620281273131,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0046262028127313105,
|
| 14 |
+
"grad_norm": 0.4456588923931122,
|
| 15 |
+
"learning_rate": 0.00019909326424870466,
|
| 16 |
+
"loss": 0.8481,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.009252405625462621,
|
| 21 |
+
"grad_norm": 0.36097365617752075,
|
| 22 |
+
"learning_rate": 0.0001981680236861584,
|
| 23 |
+
"loss": 0.6286,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.01387860843819393,
|
| 28 |
+
"grad_norm": 0.42441412806510925,
|
| 29 |
+
"learning_rate": 0.00019724278312361216,
|
| 30 |
+
"loss": 0.6207,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.018504811250925242,
|
| 35 |
+
"grad_norm": 0.42437243461608887,
|
| 36 |
+
"learning_rate": 0.0001963175425610659,
|
| 37 |
+
"loss": 0.5968,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02313101406365655,
|
| 42 |
+
"grad_norm": 0.39017853140830994,
|
| 43 |
+
"learning_rate": 0.00019539230199851963,
|
| 44 |
+
"loss": 0.6014,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02775721687638786,
|
| 49 |
+
"grad_norm": 0.39022767543792725,
|
| 50 |
+
"learning_rate": 0.00019446706143597334,
|
| 51 |
+
"loss": 0.6082,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.03238341968911917,
|
| 56 |
+
"grad_norm": 0.35285571217536926,
|
| 57 |
+
"learning_rate": 0.0001935418208734271,
|
| 58 |
+
"loss": 0.5834,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.037009622501850484,
|
| 63 |
+
"grad_norm": 0.35125309228897095,
|
| 64 |
+
"learning_rate": 0.00019261658031088084,
|
| 65 |
+
"loss": 0.5622,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.04163582531458179,
|
| 70 |
+
"grad_norm": 0.39975836873054504,
|
| 71 |
+
"learning_rate": 0.00019169133974833457,
|
| 72 |
+
"loss": 0.574,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0462620281273131,
|
| 77 |
+
"grad_norm": 0.3551945984363556,
|
| 78 |
+
"learning_rate": 0.0001907660991857883,
|
| 79 |
+
"loss": 0.5919,
|
| 80 |
+
"step": 500
|
| 81 |
+
}
|
| 82 |
+
],
|
| 83 |
+
"logging_steps": 50,
|
| 84 |
+
"max_steps": 10808,
|
| 85 |
+
"num_input_tokens_seen": 0,
|
| 86 |
+
"num_train_epochs": 1,
|
| 87 |
+
"save_steps": 500,
|
| 88 |
+
"stateful_callbacks": {
|
| 89 |
+
"TrainerControl": {
|
| 90 |
+
"args": {
|
| 91 |
+
"should_epoch_stop": false,
|
| 92 |
+
"should_evaluate": false,
|
| 93 |
+
"should_log": false,
|
| 94 |
+
"should_save": true,
|
| 95 |
+
"should_training_stop": false
|
| 96 |
+
},
|
| 97 |
+
"attributes": {}
|
| 98 |
+
}
|
| 99 |
+
},
|
| 100 |
+
"total_flos": 5.777697637023744e+16,
|
| 101 |
+
"train_batch_size": 2,
|
| 102 |
+
"trial_name": null,
|
| 103 |
+
"trial_params": null
|
| 104 |
+
}
|