Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +733 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,733 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.2810248198558847,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.012810248198558846,
|
| 13 |
+
"grad_norm": 14.422553569632921,
|
| 14 |
+
"learning_rate": 1.25e-07,
|
| 15 |
+
"loss": 0.3727,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.025620496397117692,
|
| 20 |
+
"grad_norm": 15.27369825740346,
|
| 21 |
+
"learning_rate": 2.5e-07,
|
| 22 |
+
"loss": 0.3497,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03843074459567654,
|
| 27 |
+
"grad_norm": 14.934103011301683,
|
| 28 |
+
"learning_rate": 3.75e-07,
|
| 29 |
+
"loss": 0.3837,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.051240992794235385,
|
| 34 |
+
"grad_norm": 14.736814811452579,
|
| 35 |
+
"learning_rate": 5e-07,
|
| 36 |
+
"loss": 0.3396,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06405124099279423,
|
| 41 |
+
"grad_norm": 14.197297787283581,
|
| 42 |
+
"learning_rate": 6.249999999999999e-07,
|
| 43 |
+
"loss": 0.3499,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07686148919135308,
|
| 48 |
+
"grad_norm": 11.495057131004048,
|
| 49 |
+
"learning_rate": 7.5e-07,
|
| 50 |
+
"loss": 0.3044,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08967173738991192,
|
| 55 |
+
"grad_norm": 12.10719572952739,
|
| 56 |
+
"learning_rate": 8.75e-07,
|
| 57 |
+
"loss": 0.3489,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10248198558847077,
|
| 62 |
+
"grad_norm": 17.82877298105584,
|
| 63 |
+
"learning_rate": 1e-06,
|
| 64 |
+
"loss": 0.3319,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11529223378702963,
|
| 69 |
+
"grad_norm": 20.757666788469532,
|
| 70 |
+
"learning_rate": 1.125e-06,
|
| 71 |
+
"loss": 0.299,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.12810248198558846,
|
| 76 |
+
"grad_norm": 19.06167752186461,
|
| 77 |
+
"learning_rate": 1.2499999999999999e-06,
|
| 78 |
+
"loss": 0.3083,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14091273018414732,
|
| 83 |
+
"grad_norm": 17.862857075250577,
|
| 84 |
+
"learning_rate": 1.375e-06,
|
| 85 |
+
"loss": 0.289,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15372297838270615,
|
| 90 |
+
"grad_norm": 24.872907286961055,
|
| 91 |
+
"learning_rate": 1.5e-06,
|
| 92 |
+
"loss": 0.2736,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.16653322658126501,
|
| 97 |
+
"grad_norm": 26.286295562663373,
|
| 98 |
+
"learning_rate": 1.625e-06,
|
| 99 |
+
"loss": 0.2917,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.17934347477982385,
|
| 104 |
+
"grad_norm": 27.853856612497783,
|
| 105 |
+
"learning_rate": 1.75e-06,
|
| 106 |
+
"loss": 0.3264,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.1921537229783827,
|
| 111 |
+
"grad_norm": 29.926337896366878,
|
| 112 |
+
"learning_rate": 1.8749999999999998e-06,
|
| 113 |
+
"loss": 0.4142,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.20496397117694154,
|
| 118 |
+
"grad_norm": 32.72677695738821,
|
| 119 |
+
"learning_rate": 2e-06,
|
| 120 |
+
"loss": 0.3857,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2177742193755004,
|
| 125 |
+
"grad_norm": 25.15556631940085,
|
| 126 |
+
"learning_rate": 1.9997482349425066e-06,
|
| 127 |
+
"loss": 0.4154,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.23058446757405926,
|
| 132 |
+
"grad_norm": 25.6375788037228,
|
| 133 |
+
"learning_rate": 1.9989930665413145e-06,
|
| 134 |
+
"loss": 0.393,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.2433947157726181,
|
| 139 |
+
"grad_norm": 25.369663386464367,
|
| 140 |
+
"learning_rate": 1.997734875046456e-06,
|
| 141 |
+
"loss": 0.3194,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2562049639711769,
|
| 146 |
+
"grad_norm": 22.33431342233663,
|
| 147 |
+
"learning_rate": 1.995974293995239e-06,
|
| 148 |
+
"loss": 0.3499,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2690152121697358,
|
| 153 |
+
"grad_norm": 17.384680330316037,
|
| 154 |
+
"learning_rate": 1.9937122098932426e-06,
|
| 155 |
+
"loss": 0.3306,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.28182546036829464,
|
| 160 |
+
"grad_norm": 19.864128436150512,
|
| 161 |
+
"learning_rate": 1.9909497617679347e-06,
|
| 162 |
+
"loss": 0.3596,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2946357085668535,
|
| 167 |
+
"grad_norm": 20.439398805573077,
|
| 168 |
+
"learning_rate": 1.9876883405951377e-06,
|
| 169 |
+
"loss": 0.328,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3074459567654123,
|
| 174 |
+
"grad_norm": 16.880587852141087,
|
| 175 |
+
"learning_rate": 1.9839295885986295e-06,
|
| 176 |
+
"loss": 0.2997,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.32025620496397117,
|
| 181 |
+
"grad_norm": 14.039028426941659,
|
| 182 |
+
"learning_rate": 1.9796753984232355e-06,
|
| 183 |
+
"loss": 0.2995,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.33306645316253003,
|
| 188 |
+
"grad_norm": 16.079146586971746,
|
| 189 |
+
"learning_rate": 1.9749279121818236e-06,
|
| 190 |
+
"loss": 0.3277,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3458767013610889,
|
| 195 |
+
"grad_norm": 16.540864706951407,
|
| 196 |
+
"learning_rate": 1.9696895203766866e-06,
|
| 197 |
+
"loss": 0.3492,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3586869495596477,
|
| 202 |
+
"grad_norm": 12.581391856103888,
|
| 203 |
+
"learning_rate": 1.9639628606958534e-06,
|
| 204 |
+
"loss": 0.2907,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.37149719775820655,
|
| 209 |
+
"grad_norm": 11.140674253319334,
|
| 210 |
+
"learning_rate": 1.9577508166849303e-06,
|
| 211 |
+
"loss": 0.2977,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3843074459567654,
|
| 216 |
+
"grad_norm": 14.824558664988295,
|
| 217 |
+
"learning_rate": 1.9510565162951534e-06,
|
| 218 |
+
"loss": 0.3211,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3971176941553243,
|
| 223 |
+
"grad_norm": 13.545165184398623,
|
| 224 |
+
"learning_rate": 1.9438833303083674e-06,
|
| 225 |
+
"loss": 0.3418,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4099279423538831,
|
| 230 |
+
"grad_norm": 13.298757689081272,
|
| 231 |
+
"learning_rate": 1.936234870639737e-06,
|
| 232 |
+
"loss": 0.3298,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.42273819055244194,
|
| 237 |
+
"grad_norm": 13.834770648936974,
|
| 238 |
+
"learning_rate": 1.928114988519039e-06,
|
| 239 |
+
"loss": 0.3406,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4355484387510008,
|
| 244 |
+
"grad_norm": 14.126211131938067,
|
| 245 |
+
"learning_rate": 1.9195277725514506e-06,
|
| 246 |
+
"loss": 0.3268,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.44835868694955966,
|
| 251 |
+
"grad_norm": 12.61217701807801,
|
| 252 |
+
"learning_rate": 1.9104775466588157e-06,
|
| 253 |
+
"loss": 0.3421,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4611689351481185,
|
| 258 |
+
"grad_norm": 12.638382733573478,
|
| 259 |
+
"learning_rate": 1.9009688679024189e-06,
|
| 260 |
+
"loss": 0.3311,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4739791833466773,
|
| 265 |
+
"grad_norm": 13.067396281150065,
|
| 266 |
+
"learning_rate": 1.8910065241883678e-06,
|
| 267 |
+
"loss": 0.3278,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4867894315452362,
|
| 272 |
+
"grad_norm": 12.323205100444895,
|
| 273 |
+
"learning_rate": 1.8805955318567379e-06,
|
| 274 |
+
"loss": 0.3021,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.49959967974379504,
|
| 279 |
+
"grad_norm": 15.67799328226583,
|
| 280 |
+
"learning_rate": 1.8697411331556953e-06,
|
| 281 |
+
"loss": 0.356,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5124099279423538,
|
| 286 |
+
"grad_norm": 11.446755693352566,
|
| 287 |
+
"learning_rate": 1.858448793601866e-06,
|
| 288 |
+
"loss": 0.2987,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5252201761409128,
|
| 293 |
+
"grad_norm": 13.823237028823755,
|
| 294 |
+
"learning_rate": 1.8467241992282841e-06,
|
| 295 |
+
"loss": 0.3212,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5380304243394716,
|
| 300 |
+
"grad_norm": 14.097283611242158,
|
| 301 |
+
"learning_rate": 1.8345732537213026e-06,
|
| 302 |
+
"loss": 0.2847,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5508406725380304,
|
| 307 |
+
"grad_norm": 13.339593293425303,
|
| 308 |
+
"learning_rate": 1.82200207544791e-06,
|
| 309 |
+
"loss": 0.3023,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5636509207365893,
|
| 314 |
+
"grad_norm": 11.710132196735142,
|
| 315 |
+
"learning_rate": 1.8090169943749474e-06,
|
| 316 |
+
"loss": 0.288,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5764611689351481,
|
| 321 |
+
"grad_norm": 15.421270027221475,
|
| 322 |
+
"learning_rate": 1.795624548881781e-06,
|
| 323 |
+
"loss": 0.3191,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.589271417133707,
|
| 328 |
+
"grad_norm": 15.196185783638896,
|
| 329 |
+
"learning_rate": 1.7818314824680298e-06,
|
| 330 |
+
"loss": 0.2917,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6020816653322658,
|
| 335 |
+
"grad_norm": 14.417769465666634,
|
| 336 |
+
"learning_rate": 1.767644740358011e-06,
|
| 337 |
+
"loss": 0.3065,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6148919135308246,
|
| 342 |
+
"grad_norm": 15.806685481243651,
|
| 343 |
+
"learning_rate": 1.753071466003611e-06,
|
| 344 |
+
"loss": 0.3133,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6277021617293835,
|
| 349 |
+
"grad_norm": 12.804839878435857,
|
| 350 |
+
"learning_rate": 1.7381189974873407e-06,
|
| 351 |
+
"loss": 0.2924,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6405124099279423,
|
| 356 |
+
"grad_norm": 12.41861829102855,
|
| 357 |
+
"learning_rate": 1.7227948638273915e-06,
|
| 358 |
+
"loss": 0.2884,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6533226581265013,
|
| 363 |
+
"grad_norm": 14.521843299259071,
|
| 364 |
+
"learning_rate": 1.7071067811865474e-06,
|
| 365 |
+
"loss": 0.3172,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6661329063250601,
|
| 370 |
+
"grad_norm": 12.369940546930579,
|
| 371 |
+
"learning_rate": 1.6910626489868648e-06,
|
| 372 |
+
"loss": 0.3064,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6789431545236189,
|
| 377 |
+
"grad_norm": 13.075907929491837,
|
| 378 |
+
"learning_rate": 1.6746705459320744e-06,
|
| 379 |
+
"loss": 0.3077,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6917534027221778,
|
| 384 |
+
"grad_norm": 15.91307811791754,
|
| 385 |
+
"learning_rate": 1.6579387259397126e-06,
|
| 386 |
+
"loss": 0.3202,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7045636509207366,
|
| 391 |
+
"grad_norm": 16.794702833290078,
|
| 392 |
+
"learning_rate": 1.640875613985024e-06,
|
| 393 |
+
"loss": 0.3509,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7173738991192954,
|
| 398 |
+
"grad_norm": 11.645931539680932,
|
| 399 |
+
"learning_rate": 1.6234898018587336e-06,
|
| 400 |
+
"loss": 0.2737,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7301841473178543,
|
| 405 |
+
"grad_norm": 12.164042324452254,
|
| 406 |
+
"learning_rate": 1.6057900438408199e-06,
|
| 407 |
+
"loss": 0.2805,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7429943955164131,
|
| 412 |
+
"grad_norm": 11.003522632810185,
|
| 413 |
+
"learning_rate": 1.587785252292473e-06,
|
| 414 |
+
"loss": 0.2935,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.755804643714972,
|
| 419 |
+
"grad_norm": 14.747618364345833,
|
| 420 |
+
"learning_rate": 1.569484493168452e-06,
|
| 421 |
+
"loss": 0.3185,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7686148919135308,
|
| 426 |
+
"grad_norm": 13.515142475366316,
|
| 427 |
+
"learning_rate": 1.5508969814521024e-06,
|
| 428 |
+
"loss": 0.2973,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7814251401120896,
|
| 433 |
+
"grad_norm": 12.718973924439272,
|
| 434 |
+
"learning_rate": 1.5320320765153365e-06,
|
| 435 |
+
"loss": 0.2956,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.7942353883106485,
|
| 440 |
+
"grad_norm": 11.338497570766252,
|
| 441 |
+
"learning_rate": 1.5128992774059062e-06,
|
| 442 |
+
"loss": 0.3127,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8070456365092074,
|
| 447 |
+
"grad_norm": 11.44549619969602,
|
| 448 |
+
"learning_rate": 1.4935082180643467e-06,
|
| 449 |
+
"loss": 0.277,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8198558847077662,
|
| 454 |
+
"grad_norm": 12.057954732711101,
|
| 455 |
+
"learning_rate": 1.4738686624729987e-06,
|
| 456 |
+
"loss": 0.2957,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8326661329063251,
|
| 461 |
+
"grad_norm": 12.801694035766888,
|
| 462 |
+
"learning_rate": 1.4539904997395467e-06,
|
| 463 |
+
"loss": 0.2907,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8454763811048839,
|
| 468 |
+
"grad_norm": 12.990091142294833,
|
| 469 |
+
"learning_rate": 1.433883739117558e-06,
|
| 470 |
+
"loss": 0.2973,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8582866293034428,
|
| 475 |
+
"grad_norm": 12.952175792300656,
|
| 476 |
+
"learning_rate": 1.4135585049665206e-06,
|
| 477 |
+
"loss": 0.3178,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8710968775020016,
|
| 482 |
+
"grad_norm": 12.728890379024975,
|
| 483 |
+
"learning_rate": 1.3930250316539235e-06,
|
| 484 |
+
"loss": 0.2974,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8839071257005604,
|
| 489 |
+
"grad_norm": 14.929834721351973,
|
| 490 |
+
"learning_rate": 1.3722936584019451e-06,
|
| 491 |
+
"loss": 0.2682,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.8967173738991193,
|
| 496 |
+
"grad_norm": 13.241782488686244,
|
| 497 |
+
"learning_rate": 1.3513748240813427e-06,
|
| 498 |
+
"loss": 0.3008,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9095276220976781,
|
| 503 |
+
"grad_norm": 11.993846811913796,
|
| 504 |
+
"learning_rate": 1.3302790619551672e-06,
|
| 505 |
+
"loss": 0.2954,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.922337870296237,
|
| 510 |
+
"grad_norm": 11.51482248851063,
|
| 511 |
+
"learning_rate": 1.3090169943749473e-06,
|
| 512 |
+
"loss": 0.2839,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9351481184947958,
|
| 517 |
+
"grad_norm": 16.30894659038507,
|
| 518 |
+
"learning_rate": 1.2875993274320173e-06,
|
| 519 |
+
"loss": 0.2697,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9479583666933546,
|
| 524 |
+
"grad_norm": 14.18702210008118,
|
| 525 |
+
"learning_rate": 1.266036845566675e-06,
|
| 526 |
+
"loss": 0.3257,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.9607686148919136,
|
| 531 |
+
"grad_norm": 12.829612982865,
|
| 532 |
+
"learning_rate": 1.244340406137894e-06,
|
| 533 |
+
"loss": 0.2843,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9735788630904724,
|
| 538 |
+
"grad_norm": 11.280572000591482,
|
| 539 |
+
"learning_rate": 1.2225209339563143e-06,
|
| 540 |
+
"loss": 0.2877,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.9863891112890312,
|
| 545 |
+
"grad_norm": 11.890947903730776,
|
| 546 |
+
"learning_rate": 1.2005894157832728e-06,
|
| 547 |
+
"loss": 0.3159,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.9991993594875901,
|
| 552 |
+
"grad_norm": 11.43791196956486,
|
| 553 |
+
"learning_rate": 1.1785568947986366e-06,
|
| 554 |
+
"loss": 0.2783,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.012009607686149,
|
| 559 |
+
"grad_norm": 6.708403841137654,
|
| 560 |
+
"learning_rate": 1.156434465040231e-06,
|
| 561 |
+
"loss": 0.1292,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.0248198558847077,
|
| 566 |
+
"grad_norm": 7.089840959480307,
|
| 567 |
+
"learning_rate": 1.1342332658176555e-06,
|
| 568 |
+
"loss": 0.1441,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.0376301040832665,
|
| 573 |
+
"grad_norm": 7.0309806199688705,
|
| 574 |
+
"learning_rate": 1.1119644761033077e-06,
|
| 575 |
+
"loss": 0.1405,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.0504403522818255,
|
| 580 |
+
"grad_norm": 8.710657806703308,
|
| 581 |
+
"learning_rate": 1.0896393089034335e-06,
|
| 582 |
+
"loss": 0.1265,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.0632506004803843,
|
| 587 |
+
"grad_norm": 8.365738125681913,
|
| 588 |
+
"learning_rate": 1.0672690056120398e-06,
|
| 589 |
+
"loss": 0.121,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.0760608486789431,
|
| 594 |
+
"grad_norm": 6.240558508799005,
|
| 595 |
+
"learning_rate": 1.044864830350515e-06,
|
| 596 |
+
"loss": 0.1072,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.088871096877502,
|
| 601 |
+
"grad_norm": 8.786219507218826,
|
| 602 |
+
"learning_rate": 1.022438064295805e-06,
|
| 603 |
+
"loss": 0.1256,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.1016813450760607,
|
| 608 |
+
"grad_norm": 7.367696121948762,
|
| 609 |
+
"learning_rate": 1e-06,
|
| 610 |
+
"loss": 0.1169,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.1144915932746198,
|
| 615 |
+
"grad_norm": 8.286075087364262,
|
| 616 |
+
"learning_rate": 9.77561935704195e-07,
|
| 617 |
+
"loss": 0.1302,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.1273018414731786,
|
| 622 |
+
"grad_norm": 6.69695777215013,
|
| 623 |
+
"learning_rate": 9.551351696494853e-07,
|
| 624 |
+
"loss": 0.1191,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.1401120896717374,
|
| 629 |
+
"grad_norm": 12.22193531590102,
|
| 630 |
+
"learning_rate": 9.327309943879603e-07,
|
| 631 |
+
"loss": 0.1089,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.1529223378702962,
|
| 636 |
+
"grad_norm": 9.221889701081018,
|
| 637 |
+
"learning_rate": 9.103606910965665e-07,
|
| 638 |
+
"loss": 0.1224,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.165732586068855,
|
| 643 |
+
"grad_norm": 19.718305211304475,
|
| 644 |
+
"learning_rate": 8.880355238966921e-07,
|
| 645 |
+
"loss": 0.1211,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.178542834267414,
|
| 650 |
+
"grad_norm": 8.925338055742081,
|
| 651 |
+
"learning_rate": 8.657667341823448e-07,
|
| 652 |
+
"loss": 0.1178,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.1913530824659728,
|
| 657 |
+
"grad_norm": 8.25397448128133,
|
| 658 |
+
"learning_rate": 8.435655349597689e-07,
|
| 659 |
+
"loss": 0.127,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.2041633306645316,
|
| 664 |
+
"grad_norm": 10.64375599591567,
|
| 665 |
+
"learning_rate": 8.214431052013634e-07,
|
| 666 |
+
"loss": 0.1162,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.2169735788630904,
|
| 671 |
+
"grad_norm": 9.047517084249554,
|
| 672 |
+
"learning_rate": 7.994105842167272e-07,
|
| 673 |
+
"loss": 0.1386,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.2297838270616492,
|
| 678 |
+
"grad_norm": 14.178667620453645,
|
| 679 |
+
"learning_rate": 7.774790660436857e-07,
|
| 680 |
+
"loss": 0.1137,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.2425940752602083,
|
| 685 |
+
"grad_norm": 7.4649396301693836,
|
| 686 |
+
"learning_rate": 7.556595938621058e-07,
|
| 687 |
+
"loss": 0.1189,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.255404323458767,
|
| 692 |
+
"grad_norm": 9.120043718229002,
|
| 693 |
+
"learning_rate": 7.33963154433325e-07,
|
| 694 |
+
"loss": 0.1218,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.2682145716573259,
|
| 699 |
+
"grad_norm": 9.320538317849792,
|
| 700 |
+
"learning_rate": 7.124006725679828e-07,
|
| 701 |
+
"loss": 0.1215,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.2810248198558847,
|
| 706 |
+
"grad_norm": 8.775423616734841,
|
| 707 |
+
"learning_rate": 6.909830056250526e-07,
|
| 708 |
+
"loss": 0.1106,
|
| 709 |
+
"step": 100
|
| 710 |
+
}
|
| 711 |
+
],
|
| 712 |
+
"logging_steps": 1,
|
| 713 |
+
"max_steps": 156,
|
| 714 |
+
"num_input_tokens_seen": 0,
|
| 715 |
+
"num_train_epochs": 2,
|
| 716 |
+
"save_steps": 100,
|
| 717 |
+
"stateful_callbacks": {
|
| 718 |
+
"TrainerControl": {
|
| 719 |
+
"args": {
|
| 720 |
+
"should_epoch_stop": false,
|
| 721 |
+
"should_evaluate": false,
|
| 722 |
+
"should_log": false,
|
| 723 |
+
"should_save": true,
|
| 724 |
+
"should_training_stop": false
|
| 725 |
+
},
|
| 726 |
+
"attributes": {}
|
| 727 |
+
}
|
| 728 |
+
},
|
| 729 |
+
"total_flos": 135343586148352.0,
|
| 730 |
+
"train_batch_size": 2,
|
| 731 |
+
"trial_name": null,
|
| 732 |
+
"trial_params": null
|
| 733 |
+
}
|