Upload checkpoint-104/trainer_state.json with huggingface_hub
Browse files
checkpoint-104/trainer_state.json
ADDED
|
@@ -0,0 +1,905 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 6,
|
| 6 |
+
"global_step": 104,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.038461538461538464,
|
| 13 |
+
"grad_norm": 8.71969985961914,
|
| 14 |
+
"learning_rate": 1e-05,
|
| 15 |
+
"loss": 1.7925,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.038461538461538464,
|
| 20 |
+
"eval_loss": 2.0411531925201416,
|
| 21 |
+
"eval_runtime": 1.6566,
|
| 22 |
+
"eval_samples_per_second": 13.28,
|
| 23 |
+
"eval_steps_per_second": 1.207,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.07692307692307693,
|
| 28 |
+
"grad_norm": 4.587398052215576,
|
| 29 |
+
"learning_rate": 2e-05,
|
| 30 |
+
"loss": 2.1894,
|
| 31 |
+
"step": 2
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.11538461538461539,
|
| 35 |
+
"grad_norm": 4.494429111480713,
|
| 36 |
+
"learning_rate": 3e-05,
|
| 37 |
+
"loss": 1.9438,
|
| 38 |
+
"step": 3
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15384615384615385,
|
| 42 |
+
"grad_norm": 7.78571891784668,
|
| 43 |
+
"learning_rate": 4e-05,
|
| 44 |
+
"loss": 1.9463,
|
| 45 |
+
"step": 4
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.19230769230769232,
|
| 49 |
+
"grad_norm": 6.792930603027344,
|
| 50 |
+
"learning_rate": 5e-05,
|
| 51 |
+
"loss": 1.8652,
|
| 52 |
+
"step": 5
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.23076923076923078,
|
| 56 |
+
"grad_norm": 10.765946388244629,
|
| 57 |
+
"learning_rate": 6e-05,
|
| 58 |
+
"loss": 1.6872,
|
| 59 |
+
"step": 6
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.23076923076923078,
|
| 63 |
+
"eval_loss": 1.6089171171188354,
|
| 64 |
+
"eval_runtime": 1.667,
|
| 65 |
+
"eval_samples_per_second": 13.198,
|
| 66 |
+
"eval_steps_per_second": 1.2,
|
| 67 |
+
"step": 6
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.2692307692307692,
|
| 71 |
+
"grad_norm": 3.0110535621643066,
|
| 72 |
+
"learning_rate": 7e-05,
|
| 73 |
+
"loss": 1.6877,
|
| 74 |
+
"step": 7
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.3076923076923077,
|
| 78 |
+
"grad_norm": 2.871809959411621,
|
| 79 |
+
"learning_rate": 8e-05,
|
| 80 |
+
"loss": 1.4855,
|
| 81 |
+
"step": 8
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.34615384615384615,
|
| 85 |
+
"grad_norm": 2.223036050796509,
|
| 86 |
+
"learning_rate": 9e-05,
|
| 87 |
+
"loss": 1.2108,
|
| 88 |
+
"step": 9
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.38461538461538464,
|
| 92 |
+
"grad_norm": 2.2891507148742676,
|
| 93 |
+
"learning_rate": 0.0001,
|
| 94 |
+
"loss": 1.0259,
|
| 95 |
+
"step": 10
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.4230769230769231,
|
| 99 |
+
"grad_norm": 2.058546543121338,
|
| 100 |
+
"learning_rate": 9.997207818651274e-05,
|
| 101 |
+
"loss": 0.8116,
|
| 102 |
+
"step": 11
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.46153846153846156,
|
| 106 |
+
"grad_norm": 2.8107495307922363,
|
| 107 |
+
"learning_rate": 9.988834393115767e-05,
|
| 108 |
+
"loss": 0.6967,
|
| 109 |
+
"step": 12
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.46153846153846156,
|
| 113 |
+
"eval_loss": 0.6327503323554993,
|
| 114 |
+
"eval_runtime": 1.6315,
|
| 115 |
+
"eval_samples_per_second": 13.485,
|
| 116 |
+
"eval_steps_per_second": 1.226,
|
| 117 |
+
"step": 12
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.5,
|
| 121 |
+
"grad_norm": 2.0721328258514404,
|
| 122 |
+
"learning_rate": 9.974889075442521e-05,
|
| 123 |
+
"loss": 0.5983,
|
| 124 |
+
"step": 13
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.5384615384615384,
|
| 128 |
+
"grad_norm": 1.9303380250930786,
|
| 129 |
+
"learning_rate": 9.9553874407739e-05,
|
| 130 |
+
"loss": 0.4963,
|
| 131 |
+
"step": 14
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.5769230769230769,
|
| 135 |
+
"grad_norm": 1.40817391872406,
|
| 136 |
+
"learning_rate": 9.930351269950143e-05,
|
| 137 |
+
"loss": 0.3881,
|
| 138 |
+
"step": 15
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.6153846153846154,
|
| 142 |
+
"grad_norm": 1.4169831275939941,
|
| 143 |
+
"learning_rate": 9.899808525182935e-05,
|
| 144 |
+
"loss": 0.3139,
|
| 145 |
+
"step": 16
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.6538461538461539,
|
| 149 |
+
"grad_norm": 0.9847315549850464,
|
| 150 |
+
"learning_rate": 9.863793318825186e-05,
|
| 151 |
+
"loss": 0.2776,
|
| 152 |
+
"step": 17
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.6923076923076923,
|
| 156 |
+
"grad_norm": 1.2575572729110718,
|
| 157 |
+
"learning_rate": 9.822345875271883e-05,
|
| 158 |
+
"loss": 0.3327,
|
| 159 |
+
"step": 18
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.6923076923076923,
|
| 163 |
+
"eval_loss": 0.2710803747177124,
|
| 164 |
+
"eval_runtime": 1.6739,
|
| 165 |
+
"eval_samples_per_second": 13.143,
|
| 166 |
+
"eval_steps_per_second": 1.195,
|
| 167 |
+
"step": 18
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.7307692307692307,
|
| 171 |
+
"grad_norm": 0.7373743057250977,
|
| 172 |
+
"learning_rate": 9.775512486034563e-05,
|
| 173 |
+
"loss": 0.1967,
|
| 174 |
+
"step": 19
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.7692307692307693,
|
| 178 |
+
"grad_norm": 0.8841055035591125,
|
| 179 |
+
"learning_rate": 9.723345458039594e-05,
|
| 180 |
+
"loss": 0.244,
|
| 181 |
+
"step": 20
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.8076923076923077,
|
| 185 |
+
"grad_norm": 0.8621382713317871,
|
| 186 |
+
"learning_rate": 9.665903055208014e-05,
|
| 187 |
+
"loss": 0.1927,
|
| 188 |
+
"step": 21
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.8461538461538461,
|
| 192 |
+
"grad_norm": 0.9871663451194763,
|
| 193 |
+
"learning_rate": 9.603249433382144e-05,
|
| 194 |
+
"loss": 0.2435,
|
| 195 |
+
"step": 22
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.8846153846153846,
|
| 199 |
+
"grad_norm": 0.6780880689620972,
|
| 200 |
+
"learning_rate": 9.535454568671704e-05,
|
| 201 |
+
"loss": 0.1498,
|
| 202 |
+
"step": 23
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.9230769230769231,
|
| 206 |
+
"grad_norm": 0.7006078958511353,
|
| 207 |
+
"learning_rate": 9.462594179299406e-05,
|
| 208 |
+
"loss": 0.1784,
|
| 209 |
+
"step": 24
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.9230769230769231,
|
| 213 |
+
"eval_loss": 0.17325076460838318,
|
| 214 |
+
"eval_runtime": 1.6307,
|
| 215 |
+
"eval_samples_per_second": 13.491,
|
| 216 |
+
"eval_steps_per_second": 1.226,
|
| 217 |
+
"step": 24
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.9615384615384616,
|
| 221 |
+
"grad_norm": 0.69418865442276,
|
| 222 |
+
"learning_rate": 9.384749641033359e-05,
|
| 223 |
+
"loss": 0.2146,
|
| 224 |
+
"step": 25
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.0,
|
| 228 |
+
"grad_norm": 0.5629783272743225,
|
| 229 |
+
"learning_rate": 9.302007896300698e-05,
|
| 230 |
+
"loss": 0.1474,
|
| 231 |
+
"step": 26
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.0384615384615385,
|
| 235 |
+
"grad_norm": 0.5029877424240112,
|
| 236 |
+
"learning_rate": 9.214461357083985e-05,
|
| 237 |
+
"loss": 0.1358,
|
| 238 |
+
"step": 27
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.0769230769230769,
|
| 242 |
+
"grad_norm": 0.5899357199668884,
|
| 243 |
+
"learning_rate": 9.122207801708802e-05,
|
| 244 |
+
"loss": 0.134,
|
| 245 |
+
"step": 28
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.1153846153846154,
|
| 249 |
+
"grad_norm": 0.7418510913848877,
|
| 250 |
+
"learning_rate": 9.025350265637815e-05,
|
| 251 |
+
"loss": 0.1158,
|
| 252 |
+
"step": 29
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.1538461538461537,
|
| 256 |
+
"grad_norm": 0.7511109113693237,
|
| 257 |
+
"learning_rate": 8.923996926393305e-05,
|
| 258 |
+
"loss": 0.1136,
|
| 259 |
+
"step": 30
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 1.1538461538461537,
|
| 263 |
+
"eval_loss": 0.1189715564250946,
|
| 264 |
+
"eval_runtime": 1.672,
|
| 265 |
+
"eval_samples_per_second": 13.158,
|
| 266 |
+
"eval_steps_per_second": 1.196,
|
| 267 |
+
"step": 30
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.1923076923076923,
|
| 271 |
+
"grad_norm": 0.7893090844154358,
|
| 272 |
+
"learning_rate": 8.818260982736661e-05,
|
| 273 |
+
"loss": 0.1249,
|
| 274 |
+
"step": 31
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.2307692307692308,
|
| 278 |
+
"grad_norm": 0.5709223747253418,
|
| 279 |
+
"learning_rate": 8.708260528239788e-05,
|
| 280 |
+
"loss": 0.0824,
|
| 281 |
+
"step": 32
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.2692307692307692,
|
| 285 |
+
"grad_norm": 0.5314195156097412,
|
| 286 |
+
"learning_rate": 8.594118419389647e-05,
|
| 287 |
+
"loss": 0.0906,
|
| 288 |
+
"step": 33
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.3076923076923077,
|
| 292 |
+
"grad_norm": 0.5469984412193298,
|
| 293 |
+
"learning_rate": 8.475962138373213e-05,
|
| 294 |
+
"loss": 0.0918,
|
| 295 |
+
"step": 34
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.3461538461538463,
|
| 299 |
+
"grad_norm": 0.5641104578971863,
|
| 300 |
+
"learning_rate": 8.353923650696118e-05,
|
| 301 |
+
"loss": 0.0866,
|
| 302 |
+
"step": 35
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.3846153846153846,
|
| 306 |
+
"grad_norm": 0.5818964838981628,
|
| 307 |
+
"learning_rate": 8.228139257794012e-05,
|
| 308 |
+
"loss": 0.0891,
|
| 309 |
+
"step": 36
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.3846153846153846,
|
| 313 |
+
"eval_loss": 0.08501879870891571,
|
| 314 |
+
"eval_runtime": 1.6676,
|
| 315 |
+
"eval_samples_per_second": 13.193,
|
| 316 |
+
"eval_steps_per_second": 1.199,
|
| 317 |
+
"step": 36
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.4230769230769231,
|
| 321 |
+
"grad_norm": 0.7663794159889221,
|
| 322 |
+
"learning_rate": 8.098749444801224e-05,
|
| 323 |
+
"loss": 0.0934,
|
| 324 |
+
"step": 37
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.4615384615384617,
|
| 328 |
+
"grad_norm": 0.48853516578674316,
|
| 329 |
+
"learning_rate": 7.965898723646776e-05,
|
| 330 |
+
"loss": 0.0597,
|
| 331 |
+
"step": 38
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.5,
|
| 335 |
+
"grad_norm": 0.608406662940979,
|
| 336 |
+
"learning_rate": 7.829735471652978e-05,
|
| 337 |
+
"loss": 0.0579,
|
| 338 |
+
"step": 39
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.5384615384615383,
|
| 342 |
+
"grad_norm": 0.49656111001968384,
|
| 343 |
+
"learning_rate": 7.690411765816864e-05,
|
| 344 |
+
"loss": 0.0607,
|
| 345 |
+
"step": 40
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.5769230769230769,
|
| 349 |
+
"grad_norm": 0.6709762215614319,
|
| 350 |
+
"learning_rate": 7.548083212959588e-05,
|
| 351 |
+
"loss": 0.0761,
|
| 352 |
+
"step": 41
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.6153846153846154,
|
| 356 |
+
"grad_norm": 0.6458861827850342,
|
| 357 |
+
"learning_rate": 7.402908775933419e-05,
|
| 358 |
+
"loss": 0.0746,
|
| 359 |
+
"step": 42
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.6153846153846154,
|
| 363 |
+
"eval_loss": 0.06257763504981995,
|
| 364 |
+
"eval_runtime": 1.6397,
|
| 365 |
+
"eval_samples_per_second": 13.417,
|
| 366 |
+
"eval_steps_per_second": 1.22,
|
| 367 |
+
"step": 42
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.6538461538461537,
|
| 371 |
+
"grad_norm": 0.527205765247345,
|
| 372 |
+
"learning_rate": 7.255050596080509e-05,
|
| 373 |
+
"loss": 0.0597,
|
| 374 |
+
"step": 43
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.6923076923076923,
|
| 378 |
+
"grad_norm": 0.4971027672290802,
|
| 379 |
+
"learning_rate": 7.104673812141675e-05,
|
| 380 |
+
"loss": 0.0574,
|
| 381 |
+
"step": 44
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.7307692307692308,
|
| 385 |
+
"grad_norm": 0.521812915802002,
|
| 386 |
+
"learning_rate": 6.951946375817474e-05,
|
| 387 |
+
"loss": 0.0611,
|
| 388 |
+
"step": 45
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.7692307692307692,
|
| 392 |
+
"grad_norm": 0.49732235074043274,
|
| 393 |
+
"learning_rate": 6.797038864187564e-05,
|
| 394 |
+
"loss": 0.0411,
|
| 395 |
+
"step": 46
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.8076923076923077,
|
| 399 |
+
"grad_norm": 0.4331294000148773,
|
| 400 |
+
"learning_rate": 6.640124289197845e-05,
|
| 401 |
+
"loss": 0.0377,
|
| 402 |
+
"step": 47
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.8461538461538463,
|
| 406 |
+
"grad_norm": 0.5265694260597229,
|
| 407 |
+
"learning_rate": 6.481377904428171e-05,
|
| 408 |
+
"loss": 0.0522,
|
| 409 |
+
"step": 48
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.8461538461538463,
|
| 413 |
+
"eval_loss": 0.046460047364234924,
|
| 414 |
+
"eval_runtime": 1.6754,
|
| 415 |
+
"eval_samples_per_second": 13.131,
|
| 416 |
+
"eval_steps_per_second": 1.194,
|
| 417 |
+
"step": 48
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.8846153846153846,
|
| 421 |
+
"grad_norm": 0.5977218747138977,
|
| 422 |
+
"learning_rate": 6.320977009356431e-05,
|
| 423 |
+
"loss": 0.0543,
|
| 424 |
+
"step": 49
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.9230769230769231,
|
| 428 |
+
"grad_norm": 0.4617597758769989,
|
| 429 |
+
"learning_rate": 6.159100751337642e-05,
|
| 430 |
+
"loss": 0.0388,
|
| 431 |
+
"step": 50
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.9615384615384617,
|
| 435 |
+
"grad_norm": 0.4888545572757721,
|
| 436 |
+
"learning_rate": 5.99592992551918e-05,
|
| 437 |
+
"loss": 0.0377,
|
| 438 |
+
"step": 51
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.0,
|
| 442 |
+
"grad_norm": 0.43416959047317505,
|
| 443 |
+
"learning_rate": 5.831646772915651e-05,
|
| 444 |
+
"loss": 0.0326,
|
| 445 |
+
"step": 52
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.0384615384615383,
|
| 449 |
+
"grad_norm": 0.4446072578430176,
|
| 450 |
+
"learning_rate": 5.666434776868895e-05,
|
| 451 |
+
"loss": 0.03,
|
| 452 |
+
"step": 53
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.076923076923077,
|
| 456 |
+
"grad_norm": 0.569553792476654,
|
| 457 |
+
"learning_rate": 5.5004784581204927e-05,
|
| 458 |
+
"loss": 0.033,
|
| 459 |
+
"step": 54
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.076923076923077,
|
| 463 |
+
"eval_loss": 0.028210198506712914,
|
| 464 |
+
"eval_runtime": 1.7415,
|
| 465 |
+
"eval_samples_per_second": 12.633,
|
| 466 |
+
"eval_steps_per_second": 1.148,
|
| 467 |
+
"step": 54
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.1153846153846154,
|
| 471 |
+
"grad_norm": 0.4060160219669342,
|
| 472 |
+
"learning_rate": 5.3339631687256084e-05,
|
| 473 |
+
"loss": 0.0245,
|
| 474 |
+
"step": 55
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.1538461538461537,
|
| 478 |
+
"grad_norm": 0.48645609617233276,
|
| 479 |
+
"learning_rate": 5.167074885038373e-05,
|
| 480 |
+
"loss": 0.024,
|
| 481 |
+
"step": 56
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 2.1923076923076925,
|
| 485 |
+
"grad_norm": 0.5603619813919067,
|
| 486 |
+
"learning_rate": 5e-05,
|
| 487 |
+
"loss": 0.0257,
|
| 488 |
+
"step": 57
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 2.230769230769231,
|
| 492 |
+
"grad_norm": 0.43165791034698486,
|
| 493 |
+
"learning_rate": 4.832925114961629e-05,
|
| 494 |
+
"loss": 0.0207,
|
| 495 |
+
"step": 58
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 2.269230769230769,
|
| 499 |
+
"grad_norm": 0.48327693343162537,
|
| 500 |
+
"learning_rate": 4.666036831274392e-05,
|
| 501 |
+
"loss": 0.0308,
|
| 502 |
+
"step": 59
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 2.3076923076923075,
|
| 506 |
+
"grad_norm": 0.5035133957862854,
|
| 507 |
+
"learning_rate": 4.4995215418795085e-05,
|
| 508 |
+
"loss": 0.0333,
|
| 509 |
+
"step": 60
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 2.3076923076923075,
|
| 513 |
+
"eval_loss": 0.02246815524995327,
|
| 514 |
+
"eval_runtime": 1.6386,
|
| 515 |
+
"eval_samples_per_second": 13.426,
|
| 516 |
+
"eval_steps_per_second": 1.221,
|
| 517 |
+
"step": 60
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 2.3461538461538463,
|
| 521 |
+
"grad_norm": 0.38951441645622253,
|
| 522 |
+
"learning_rate": 4.333565223131107e-05,
|
| 523 |
+
"loss": 0.018,
|
| 524 |
+
"step": 61
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.3846153846153846,
|
| 528 |
+
"grad_norm": 0.5282712578773499,
|
| 529 |
+
"learning_rate": 4.1683532270843504e-05,
|
| 530 |
+
"loss": 0.0246,
|
| 531 |
+
"step": 62
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 2.423076923076923,
|
| 535 |
+
"grad_norm": 0.29735180735588074,
|
| 536 |
+
"learning_rate": 4.0040700744808204e-05,
|
| 537 |
+
"loss": 0.0165,
|
| 538 |
+
"step": 63
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 2.4615384615384617,
|
| 542 |
+
"grad_norm": 0.3622055649757385,
|
| 543 |
+
"learning_rate": 3.840899248662358e-05,
|
| 544 |
+
"loss": 0.0203,
|
| 545 |
+
"step": 64
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 2.5,
|
| 549 |
+
"grad_norm": 0.39045998454093933,
|
| 550 |
+
"learning_rate": 3.6790229906435705e-05,
|
| 551 |
+
"loss": 0.0197,
|
| 552 |
+
"step": 65
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.5384615384615383,
|
| 556 |
+
"grad_norm": 0.39309993386268616,
|
| 557 |
+
"learning_rate": 3.5186220955718306e-05,
|
| 558 |
+
"loss": 0.0171,
|
| 559 |
+
"step": 66
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 2.5384615384615383,
|
| 563 |
+
"eval_loss": 0.020282387733459473,
|
| 564 |
+
"eval_runtime": 1.6665,
|
| 565 |
+
"eval_samples_per_second": 13.202,
|
| 566 |
+
"eval_steps_per_second": 1.2,
|
| 567 |
+
"step": 66
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.5769230769230766,
|
| 571 |
+
"grad_norm": 0.3418659269809723,
|
| 572 |
+
"learning_rate": 3.3598757108021546e-05,
|
| 573 |
+
"loss": 0.0112,
|
| 574 |
+
"step": 67
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 2.6153846153846154,
|
| 578 |
+
"grad_norm": 0.4300234317779541,
|
| 579 |
+
"learning_rate": 3.202961135812437e-05,
|
| 580 |
+
"loss": 0.018,
|
| 581 |
+
"step": 68
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 2.6538461538461537,
|
| 585 |
+
"grad_norm": 0.34998640418052673,
|
| 586 |
+
"learning_rate": 3.0480536241825263e-05,
|
| 587 |
+
"loss": 0.014,
|
| 588 |
+
"step": 69
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 2.6923076923076925,
|
| 592 |
+
"grad_norm": 0.46385401487350464,
|
| 593 |
+
"learning_rate": 2.895326187858326e-05,
|
| 594 |
+
"loss": 0.0178,
|
| 595 |
+
"step": 70
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 2.730769230769231,
|
| 599 |
+
"grad_norm": 0.4280989468097687,
|
| 600 |
+
"learning_rate": 2.74494940391949e-05,
|
| 601 |
+
"loss": 0.0195,
|
| 602 |
+
"step": 71
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 2.769230769230769,
|
| 606 |
+
"grad_norm": 0.337554395198822,
|
| 607 |
+
"learning_rate": 2.5970912240665813e-05,
|
| 608 |
+
"loss": 0.0172,
|
| 609 |
+
"step": 72
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.769230769230769,
|
| 613 |
+
"eval_loss": 0.014381876215338707,
|
| 614 |
+
"eval_runtime": 1.6934,
|
| 615 |
+
"eval_samples_per_second": 12.991,
|
| 616 |
+
"eval_steps_per_second": 1.181,
|
| 617 |
+
"step": 72
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 2.8076923076923075,
|
| 621 |
+
"grad_norm": 0.32722559571266174,
|
| 622 |
+
"learning_rate": 2.4519167870404125e-05,
|
| 623 |
+
"loss": 0.0172,
|
| 624 |
+
"step": 73
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.8461538461538463,
|
| 628 |
+
"grad_norm": 0.3325071334838867,
|
| 629 |
+
"learning_rate": 2.3095882341831372e-05,
|
| 630 |
+
"loss": 0.0113,
|
| 631 |
+
"step": 74
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.8846153846153846,
|
| 635 |
+
"grad_norm": 0.31768810749053955,
|
| 636 |
+
"learning_rate": 2.1702645283470236e-05,
|
| 637 |
+
"loss": 0.0119,
|
| 638 |
+
"step": 75
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 2.9230769230769234,
|
| 642 |
+
"grad_norm": 0.29802849888801575,
|
| 643 |
+
"learning_rate": 2.0341012763532243e-05,
|
| 644 |
+
"loss": 0.0131,
|
| 645 |
+
"step": 76
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 2.9615384615384617,
|
| 649 |
+
"grad_norm": 0.26436880230903625,
|
| 650 |
+
"learning_rate": 1.9012505551987765e-05,
|
| 651 |
+
"loss": 0.011,
|
| 652 |
+
"step": 77
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 3.0,
|
| 656 |
+
"grad_norm": 0.24293173849582672,
|
| 657 |
+
"learning_rate": 1.771860742205988e-05,
|
| 658 |
+
"loss": 0.0095,
|
| 659 |
+
"step": 78
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 3.0,
|
| 663 |
+
"eval_loss": 0.011883563362061977,
|
| 664 |
+
"eval_runtime": 1.6658,
|
| 665 |
+
"eval_samples_per_second": 13.207,
|
| 666 |
+
"eval_steps_per_second": 1.201,
|
| 667 |
+
"step": 78
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 3.0384615384615383,
|
| 671 |
+
"grad_norm": 0.19049152731895447,
|
| 672 |
+
"learning_rate": 1.646076349303884e-05,
|
| 673 |
+
"loss": 0.0094,
|
| 674 |
+
"step": 79
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 3.076923076923077,
|
| 678 |
+
"grad_norm": 0.3024927079677582,
|
| 679 |
+
"learning_rate": 1.5240378616267886e-05,
|
| 680 |
+
"loss": 0.0094,
|
| 681 |
+
"step": 80
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 3.1153846153846154,
|
| 685 |
+
"grad_norm": 0.28958773612976074,
|
| 686 |
+
"learning_rate": 1.4058815806103542e-05,
|
| 687 |
+
"loss": 0.0089,
|
| 688 |
+
"step": 81
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 3.1538461538461537,
|
| 692 |
+
"grad_norm": 0.22336073219776154,
|
| 693 |
+
"learning_rate": 1.2917394717602121e-05,
|
| 694 |
+
"loss": 0.0075,
|
| 695 |
+
"step": 82
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 3.1923076923076925,
|
| 699 |
+
"grad_norm": 0.17351990938186646,
|
| 700 |
+
"learning_rate": 1.1817390172633403e-05,
|
| 701 |
+
"loss": 0.0068,
|
| 702 |
+
"step": 83
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 3.230769230769231,
|
| 706 |
+
"grad_norm": 0.2886107563972473,
|
| 707 |
+
"learning_rate": 1.0760030736066951e-05,
|
| 708 |
+
"loss": 0.0088,
|
| 709 |
+
"step": 84
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 3.230769230769231,
|
| 713 |
+
"eval_loss": 0.009916023351252079,
|
| 714 |
+
"eval_runtime": 1.6701,
|
| 715 |
+
"eval_samples_per_second": 13.173,
|
| 716 |
+
"eval_steps_per_second": 1.198,
|
| 717 |
+
"step": 84
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 3.269230769230769,
|
| 721 |
+
"grad_norm": 0.31455695629119873,
|
| 722 |
+
"learning_rate": 9.746497343621857e-06,
|
| 723 |
+
"loss": 0.011,
|
| 724 |
+
"step": 85
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 3.3076923076923075,
|
| 728 |
+
"grad_norm": 0.22167526185512543,
|
| 729 |
+
"learning_rate": 8.777921982911996e-06,
|
| 730 |
+
"loss": 0.007,
|
| 731 |
+
"step": 86
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 3.3461538461538463,
|
| 735 |
+
"grad_norm": 0.2428387552499771,
|
| 736 |
+
"learning_rate": 7.85538642916015e-06,
|
| 737 |
+
"loss": 0.0075,
|
| 738 |
+
"step": 87
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 3.3846153846153846,
|
| 742 |
+
"grad_norm": 0.2719491720199585,
|
| 743 |
+
"learning_rate": 6.979921036993042e-06,
|
| 744 |
+
"loss": 0.0063,
|
| 745 |
+
"step": 88
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 3.423076923076923,
|
| 749 |
+
"grad_norm": 0.20811867713928223,
|
| 750 |
+
"learning_rate": 6.152503589666425e-06,
|
| 751 |
+
"loss": 0.0069,
|
| 752 |
+
"step": 89
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 3.4615384615384617,
|
| 756 |
+
"grad_norm": 0.1949089914560318,
|
| 757 |
+
"learning_rate": 5.374058207005944e-06,
|
| 758 |
+
"loss": 0.0054,
|
| 759 |
+
"step": 90
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 3.4615384615384617,
|
| 763 |
+
"eval_loss": 0.008881751447916031,
|
| 764 |
+
"eval_runtime": 1.7354,
|
| 765 |
+
"eval_samples_per_second": 12.678,
|
| 766 |
+
"eval_steps_per_second": 1.153,
|
| 767 |
+
"step": 90
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 3.5,
|
| 771 |
+
"grad_norm": 0.2331872284412384,
|
| 772 |
+
"learning_rate": 4.645454313282965e-06,
|
| 773 |
+
"loss": 0.0055,
|
| 774 |
+
"step": 91
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.5384615384615383,
|
| 778 |
+
"grad_norm": 0.19056551158428192,
|
| 779 |
+
"learning_rate": 3.967505666178556e-06,
|
| 780 |
+
"loss": 0.0058,
|
| 781 |
+
"step": 92
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 3.5769230769230766,
|
| 785 |
+
"grad_norm": 0.2065887749195099,
|
| 786 |
+
"learning_rate": 3.340969447919873e-06,
|
| 787 |
+
"loss": 0.0046,
|
| 788 |
+
"step": 93
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 3.6153846153846154,
|
| 792 |
+
"grad_norm": 0.2241547554731369,
|
| 793 |
+
"learning_rate": 2.7665454196040664e-06,
|
| 794 |
+
"loss": 0.0056,
|
| 795 |
+
"step": 94
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 3.6538461538461537,
|
| 799 |
+
"grad_norm": 0.1925225704908371,
|
| 800 |
+
"learning_rate": 2.2448751396543787e-06,
|
| 801 |
+
"loss": 0.0055,
|
| 802 |
+
"step": 95
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 3.6923076923076925,
|
| 806 |
+
"grad_norm": 0.292306125164032,
|
| 807 |
+
"learning_rate": 1.7765412472811771e-06,
|
| 808 |
+
"loss": 0.0073,
|
| 809 |
+
"step": 96
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 3.6923076923076925,
|
| 813 |
+
"eval_loss": 0.008542757481336594,
|
| 814 |
+
"eval_runtime": 1.667,
|
| 815 |
+
"eval_samples_per_second": 13.198,
|
| 816 |
+
"eval_steps_per_second": 1.2,
|
| 817 |
+
"step": 96
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 3.730769230769231,
|
| 821 |
+
"grad_norm": 0.25909164547920227,
|
| 822 |
+
"learning_rate": 1.3620668117481472e-06,
|
| 823 |
+
"loss": 0.0062,
|
| 824 |
+
"step": 97
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 3.769230769230769,
|
| 828 |
+
"grad_norm": 0.2407168447971344,
|
| 829 |
+
"learning_rate": 1.0019147481706625e-06,
|
| 830 |
+
"loss": 0.0071,
|
| 831 |
+
"step": 98
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 3.8076923076923075,
|
| 835 |
+
"grad_norm": 0.2027008980512619,
|
| 836 |
+
"learning_rate": 6.964873004985717e-07,
|
| 837 |
+
"loss": 0.009,
|
| 838 |
+
"step": 99
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 3.8461538461538463,
|
| 842 |
+
"grad_norm": 0.2740994691848755,
|
| 843 |
+
"learning_rate": 4.461255922609986e-07,
|
| 844 |
+
"loss": 0.0061,
|
| 845 |
+
"step": 100
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 3.8846153846153846,
|
| 849 |
+
"grad_norm": 0.26183342933654785,
|
| 850 |
+
"learning_rate": 2.511092455747932e-07,
|
| 851 |
+
"loss": 0.0082,
|
| 852 |
+
"step": 101
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 3.9230769230769234,
|
| 856 |
+
"grad_norm": 0.26681411266326904,
|
| 857 |
+
"learning_rate": 1.1165606884234181e-07,
|
| 858 |
+
"loss": 0.0059,
|
| 859 |
+
"step": 102
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 3.9230769230769234,
|
| 863 |
+
"eval_loss": 0.0077163358218967915,
|
| 864 |
+
"eval_runtime": 1.7901,
|
| 865 |
+
"eval_samples_per_second": 12.29,
|
| 866 |
+
"eval_steps_per_second": 1.117,
|
| 867 |
+
"step": 102
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 3.9615384615384617,
|
| 871 |
+
"grad_norm": 0.2794024646282196,
|
| 872 |
+
"learning_rate": 2.792181348726941e-08,
|
| 873 |
+
"loss": 0.0066,
|
| 874 |
+
"step": 103
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 4.0,
|
| 878 |
+
"grad_norm": 0.18571990728378296,
|
| 879 |
+
"learning_rate": 0.0,
|
| 880 |
+
"loss": 0.0063,
|
| 881 |
+
"step": 104
|
| 882 |
+
}
|
| 883 |
+
],
|
| 884 |
+
"logging_steps": 1,
|
| 885 |
+
"max_steps": 104,
|
| 886 |
+
"num_input_tokens_seen": 0,
|
| 887 |
+
"num_train_epochs": 4,
|
| 888 |
+
"save_steps": 500,
|
| 889 |
+
"stateful_callbacks": {
|
| 890 |
+
"TrainerControl": {
|
| 891 |
+
"args": {
|
| 892 |
+
"should_epoch_stop": false,
|
| 893 |
+
"should_evaluate": false,
|
| 894 |
+
"should_log": false,
|
| 895 |
+
"should_save": true,
|
| 896 |
+
"should_training_stop": true
|
| 897 |
+
},
|
| 898 |
+
"attributes": {}
|
| 899 |
+
}
|
| 900 |
+
},
|
| 901 |
+
"total_flos": 5.395758414666138e+16,
|
| 902 |
+
"train_batch_size": 4,
|
| 903 |
+
"trial_name": null,
|
| 904 |
+
"trial_params": null
|
| 905 |
+
}
|