Upload aetherargus-lora/trainer_state.json with huggingface_hub
Browse files
aetherargus-lora/trainer_state.json
ADDED
|
@@ -0,0 +1,158 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 60,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.016701461377870562,
|
| 14 |
+
"grad_norm": 4.485437870025635,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 2.8465373516082764,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.08350730688935282,
|
| 21 |
+
"grad_norm": 3.0602867603302,
|
| 22 |
+
"learning_rate": 9.992407582166581e-05,
|
| 23 |
+
"loss": 2.294060707092285,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.16701461377870563,
|
| 28 |
+
"grad_norm": 0.8693507313728333,
|
| 29 |
+
"learning_rate": 9.729086208503174e-05,
|
| 30 |
+
"loss": 1.2953631401062011,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.25052192066805845,
|
| 35 |
+
"grad_norm": 0.7621045708656311,
|
| 36 |
+
"learning_rate": 9.108889076126226e-05,
|
| 37 |
+
"loss": 0.8703152656555175,
|
| 38 |
+
"step": 15
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.33402922755741127,
|
| 42 |
+
"grad_norm": 0.6791566014289856,
|
| 43 |
+
"learning_rate": 8.178618741049842e-05,
|
| 44 |
+
"loss": 0.6290188312530518,
|
| 45 |
+
"step": 20
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.4175365344467641,
|
| 49 |
+
"grad_norm": 0.6321083307266235,
|
| 50 |
+
"learning_rate": 7.008477123264848e-05,
|
| 51 |
+
"loss": 0.46433124542236326,
|
| 52 |
+
"step": 25
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.5010438413361169,
|
| 56 |
+
"grad_norm": 0.49947911500930786,
|
| 57 |
+
"learning_rate": 5.686767789092041e-05,
|
| 58 |
+
"loss": 0.33226304054260253,
|
| 59 |
+
"step": 30
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.5845511482254697,
|
| 63 |
+
"grad_norm": 0.7511892914772034,
|
| 64 |
+
"learning_rate": 4.3132322109079596e-05,
|
| 65 |
+
"loss": 0.4396816730499268,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.6680584551148225,
|
| 70 |
+
"grad_norm": 0.5519866347312927,
|
| 71 |
+
"learning_rate": 2.991522876735154e-05,
|
| 72 |
+
"loss": 0.36175758838653566,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.7515657620041754,
|
| 77 |
+
"grad_norm": 0.36559605598449707,
|
| 78 |
+
"learning_rate": 1.821381258950161e-05,
|
| 79 |
+
"loss": 0.26281235218048093,
|
| 80 |
+
"step": 45
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.8350730688935282,
|
| 84 |
+
"grad_norm": 0.6063154339790344,
|
| 85 |
+
"learning_rate": 8.911109238737747e-06,
|
| 86 |
+
"loss": 0.3335791826248169,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.8350730688935282,
|
| 91 |
+
"eval_loss": 0.5060427784919739,
|
| 92 |
+
"eval_runtime": 23.6562,
|
| 93 |
+
"eval_samples_per_second": 5.157,
|
| 94 |
+
"eval_steps_per_second": 5.157,
|
| 95 |
+
"step": 50
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.918580375782881,
|
| 99 |
+
"grad_norm": 0.6971960663795471,
|
| 100 |
+
"learning_rate": 2.7091379149682685e-06,
|
| 101 |
+
"loss": 0.32030749320983887,
|
| 102 |
+
"step": 55
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.0,
|
| 106 |
+
"grad_norm": 0.6506057977676392,
|
| 107 |
+
"learning_rate": 7.59241783341913e-08,
|
| 108 |
+
"loss": 0.32254273891448976,
|
| 109 |
+
"step": 60
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.0,
|
| 113 |
+
"eval_loss": 0.5009067058563232,
|
| 114 |
+
"eval_runtime": 21.0472,
|
| 115 |
+
"eval_samples_per_second": 5.797,
|
| 116 |
+
"eval_steps_per_second": 5.797,
|
| 117 |
+
"step": 60
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.0,
|
| 121 |
+
"step": 60,
|
| 122 |
+
"total_flos": 8434343709060096.0,
|
| 123 |
+
"train_loss": 0.6697107156117758,
|
| 124 |
+
"train_runtime": 651.199,
|
| 125 |
+
"train_samples_per_second": 1.471,
|
| 126 |
+
"train_steps_per_second": 0.092
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.0,
|
| 130 |
+
"eval_loss": 0.5009067058563232,
|
| 131 |
+
"eval_runtime": 23.6746,
|
| 132 |
+
"eval_samples_per_second": 5.153,
|
| 133 |
+
"eval_steps_per_second": 5.153,
|
| 134 |
+
"step": 60
|
| 135 |
+
}
|
| 136 |
+
],
|
| 137 |
+
"logging_steps": 5,
|
| 138 |
+
"max_steps": 60,
|
| 139 |
+
"num_input_tokens_seen": 0,
|
| 140 |
+
"num_train_epochs": 1,
|
| 141 |
+
"save_steps": 50,
|
| 142 |
+
"stateful_callbacks": {
|
| 143 |
+
"TrainerControl": {
|
| 144 |
+
"args": {
|
| 145 |
+
"should_epoch_stop": false,
|
| 146 |
+
"should_evaluate": false,
|
| 147 |
+
"should_log": false,
|
| 148 |
+
"should_save": true,
|
| 149 |
+
"should_training_stop": true
|
| 150 |
+
},
|
| 151 |
+
"attributes": {}
|
| 152 |
+
}
|
| 153 |
+
},
|
| 154 |
+
"total_flos": 8434343709060096.0,
|
| 155 |
+
"train_batch_size": 1,
|
| 156 |
+
"trial_name": null,
|
| 157 |
+
"trial_params": null
|
| 158 |
+
}
|