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": 2.26628895184136,
|
| 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": 0.0226628895184136,
|
| 13 |
+
"grad_norm": 5.715946698612309,
|
| 14 |
+
"learning_rate": 3.7037037037037036e-08,
|
| 15 |
+
"loss": 1.039,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0453257790368272,
|
| 20 |
+
"grad_norm": 5.918098634610158,
|
| 21 |
+
"learning_rate": 7.407407407407407e-08,
|
| 22 |
+
"loss": 1.0345,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0679886685552408,
|
| 27 |
+
"grad_norm": 5.967358491879423,
|
| 28 |
+
"learning_rate": 1.111111111111111e-07,
|
| 29 |
+
"loss": 1.0568,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0906515580736544,
|
| 34 |
+
"grad_norm": 6.076151471056227,
|
| 35 |
+
"learning_rate": 1.4814814814814815e-07,
|
| 36 |
+
"loss": 1.0407,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.11331444759206799,
|
| 41 |
+
"grad_norm": 5.698276915195162,
|
| 42 |
+
"learning_rate": 1.8518518518518516e-07,
|
| 43 |
+
"loss": 1.0355,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1359773371104816,
|
| 48 |
+
"grad_norm": 5.524873495595531,
|
| 49 |
+
"learning_rate": 2.222222222222222e-07,
|
| 50 |
+
"loss": 1.0329,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.15864022662889518,
|
| 55 |
+
"grad_norm": 5.663139068043792,
|
| 56 |
+
"learning_rate": 2.5925925925925923e-07,
|
| 57 |
+
"loss": 1.013,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1813031161473088,
|
| 62 |
+
"grad_norm": 5.483842003291619,
|
| 63 |
+
"learning_rate": 2.962962962962963e-07,
|
| 64 |
+
"loss": 1.0285,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.20396600566572237,
|
| 69 |
+
"grad_norm": 5.501921058157795,
|
| 70 |
+
"learning_rate": 3.333333333333333e-07,
|
| 71 |
+
"loss": 1.0181,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.22662889518413598,
|
| 76 |
+
"grad_norm": 5.691661611678567,
|
| 77 |
+
"learning_rate": 3.703703703703703e-07,
|
| 78 |
+
"loss": 1.0046,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.24929178470254956,
|
| 83 |
+
"grad_norm": 5.490524973688248,
|
| 84 |
+
"learning_rate": 4.0740740740740737e-07,
|
| 85 |
+
"loss": 1.0291,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2719546742209632,
|
| 90 |
+
"grad_norm": 4.885236117260528,
|
| 91 |
+
"learning_rate": 4.444444444444444e-07,
|
| 92 |
+
"loss": 1.0084,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.29461756373937675,
|
| 97 |
+
"grad_norm": 5.256688897749667,
|
| 98 |
+
"learning_rate": 4.814814814814814e-07,
|
| 99 |
+
"loss": 0.9945,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.31728045325779036,
|
| 104 |
+
"grad_norm": 5.026023661790397,
|
| 105 |
+
"learning_rate": 5.185185185185185e-07,
|
| 106 |
+
"loss": 0.9936,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.33994334277620397,
|
| 111 |
+
"grad_norm": 4.979666180740075,
|
| 112 |
+
"learning_rate": 5.555555555555555e-07,
|
| 113 |
+
"loss": 0.9997,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3626062322946176,
|
| 118 |
+
"grad_norm": 4.741351636847691,
|
| 119 |
+
"learning_rate": 5.925925925925926e-07,
|
| 120 |
+
"loss": 0.9904,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.38526912181303113,
|
| 125 |
+
"grad_norm": 4.429638197959212,
|
| 126 |
+
"learning_rate": 6.296296296296296e-07,
|
| 127 |
+
"loss": 0.9779,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.40793201133144474,
|
| 132 |
+
"grad_norm": 4.2702651723674006,
|
| 133 |
+
"learning_rate": 6.666666666666666e-07,
|
| 134 |
+
"loss": 0.9373,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.43059490084985835,
|
| 139 |
+
"grad_norm": 4.371215055008036,
|
| 140 |
+
"learning_rate": 7.037037037037037e-07,
|
| 141 |
+
"loss": 0.9616,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.45325779036827196,
|
| 146 |
+
"grad_norm": 4.300078040900759,
|
| 147 |
+
"learning_rate": 7.407407407407406e-07,
|
| 148 |
+
"loss": 0.9581,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.47592067988668557,
|
| 153 |
+
"grad_norm": 4.242855799180736,
|
| 154 |
+
"learning_rate": 7.777777777777778e-07,
|
| 155 |
+
"loss": 0.9454,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4985835694050991,
|
| 160 |
+
"grad_norm": 3.4536592234259555,
|
| 161 |
+
"learning_rate": 8.148148148148147e-07,
|
| 162 |
+
"loss": 0.9274,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5212464589235127,
|
| 167 |
+
"grad_norm": 3.3525795982748203,
|
| 168 |
+
"learning_rate": 8.518518518518518e-07,
|
| 169 |
+
"loss": 0.8833,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5439093484419264,
|
| 174 |
+
"grad_norm": 3.110575381958802,
|
| 175 |
+
"learning_rate": 8.888888888888888e-07,
|
| 176 |
+
"loss": 0.9066,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.56657223796034,
|
| 181 |
+
"grad_norm": 3.18785930927135,
|
| 182 |
+
"learning_rate": 9.259259259259259e-07,
|
| 183 |
+
"loss": 0.8896,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5892351274787535,
|
| 188 |
+
"grad_norm": 3.0188412291205684,
|
| 189 |
+
"learning_rate": 9.629629629629628e-07,
|
| 190 |
+
"loss": 0.9068,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6118980169971672,
|
| 195 |
+
"grad_norm": 3.0072699515749344,
|
| 196 |
+
"learning_rate": 1e-06,
|
| 197 |
+
"loss": 0.8959,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6345609065155807,
|
| 202 |
+
"grad_norm": 3.050779999599616,
|
| 203 |
+
"learning_rate": 9.999560724782173e-07,
|
| 204 |
+
"loss": 0.8648,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6572237960339944,
|
| 209 |
+
"grad_norm": 3.034749793056673,
|
| 210 |
+
"learning_rate": 9.998242976313776e-07,
|
| 211 |
+
"loss": 0.8763,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6798866855524079,
|
| 216 |
+
"grad_norm": 2.6230160618361897,
|
| 217 |
+
"learning_rate": 9.996046986136508e-07,
|
| 218 |
+
"loss": 0.8439,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7025495750708215,
|
| 223 |
+
"grad_norm": 2.619746810094255,
|
| 224 |
+
"learning_rate": 9.992973140107996e-07,
|
| 225 |
+
"loss": 0.8395,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7252124645892352,
|
| 230 |
+
"grad_norm": 2.2660982887250496,
|
| 231 |
+
"learning_rate": 9.989021978333994e-07,
|
| 232 |
+
"loss": 0.8407,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7478753541076487,
|
| 237 |
+
"grad_norm": 1.92948640709938,
|
| 238 |
+
"learning_rate": 9.984194195073478e-07,
|
| 239 |
+
"loss": 0.8175,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7705382436260623,
|
| 244 |
+
"grad_norm": 1.8673042037436878,
|
| 245 |
+
"learning_rate": 9.97849063861667e-07,
|
| 246 |
+
"loss": 0.7963,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7932011331444759,
|
| 251 |
+
"grad_norm": 1.841378707582655,
|
| 252 |
+
"learning_rate": 9.971912311135967e-07,
|
| 253 |
+
"loss": 0.8177,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8158640226628895,
|
| 258 |
+
"grad_norm": 1.6212101538356403,
|
| 259 |
+
"learning_rate": 9.964460368509865e-07,
|
| 260 |
+
"loss": 0.8036,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.8385269121813032,
|
| 265 |
+
"grad_norm": 1.6148282593388759,
|
| 266 |
+
"learning_rate": 9.956136120119856e-07,
|
| 267 |
+
"loss": 0.7945,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8611898016997167,
|
| 272 |
+
"grad_norm": 1.5660870386151309,
|
| 273 |
+
"learning_rate": 9.946941028620347e-07,
|
| 274 |
+
"loss": 0.7919,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8838526912181303,
|
| 279 |
+
"grad_norm": 1.5162976532167538,
|
| 280 |
+
"learning_rate": 9.936876709681666e-07,
|
| 281 |
+
"loss": 0.7965,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.9065155807365439,
|
| 286 |
+
"grad_norm": 1.4779616090178773,
|
| 287 |
+
"learning_rate": 9.92594493170617e-07,
|
| 288 |
+
"loss": 0.7872,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.9291784702549575,
|
| 293 |
+
"grad_norm": 1.4588545367417372,
|
| 294 |
+
"learning_rate": 9.914147615517526e-07,
|
| 295 |
+
"loss": 0.7933,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.9518413597733711,
|
| 300 |
+
"grad_norm": 1.2450088034935203,
|
| 301 |
+
"learning_rate": 9.901486834023181e-07,
|
| 302 |
+
"loss": 0.7401,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9745042492917847,
|
| 307 |
+
"grad_norm": 1.1159548060929454,
|
| 308 |
+
"learning_rate": 9.887964811850157e-07,
|
| 309 |
+
"loss": 0.7496,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9971671388101983,
|
| 314 |
+
"grad_norm": 1.0418410473138606,
|
| 315 |
+
"learning_rate": 9.87358392495415e-07,
|
| 316 |
+
"loss": 0.7568,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.019830028328612,
|
| 321 |
+
"grad_norm": 2.1594760368768195,
|
| 322 |
+
"learning_rate": 9.858346700202048e-07,
|
| 323 |
+
"loss": 1.3469,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.0424929178470255,
|
| 328 |
+
"grad_norm": 0.9706954495224399,
|
| 329 |
+
"learning_rate": 9.842255814927944e-07,
|
| 330 |
+
"loss": 0.7412,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.065155807365439,
|
| 335 |
+
"grad_norm": 0.9479843401943371,
|
| 336 |
+
"learning_rate": 9.825314096462684e-07,
|
| 337 |
+
"loss": 0.712,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.0878186968838528,
|
| 342 |
+
"grad_norm": 0.8785518016295425,
|
| 343 |
+
"learning_rate": 9.807524521637102e-07,
|
| 344 |
+
"loss": 0.721,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.1104815864022664,
|
| 349 |
+
"grad_norm": 0.9083971698155864,
|
| 350 |
+
"learning_rate": 9.788890216258938e-07,
|
| 351 |
+
"loss": 0.7405,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.13314447592068,
|
| 356 |
+
"grad_norm": 0.9052818651846114,
|
| 357 |
+
"learning_rate": 9.769414454563615e-07,
|
| 358 |
+
"loss": 0.7223,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.1558073654390935,
|
| 363 |
+
"grad_norm": 0.8244297426454674,
|
| 364 |
+
"learning_rate": 9.749100658638914e-07,
|
| 365 |
+
"loss": 0.7113,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.178470254957507,
|
| 370 |
+
"grad_norm": 0.7448472800310213,
|
| 371 |
+
"learning_rate": 9.72795239782369e-07,
|
| 372 |
+
"loss": 0.7001,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.2011331444759206,
|
| 377 |
+
"grad_norm": 0.8936397991398377,
|
| 378 |
+
"learning_rate": 9.705973388080692e-07,
|
| 379 |
+
"loss": 0.6924,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.2237960339943343,
|
| 384 |
+
"grad_norm": 0.7188466048624885,
|
| 385 |
+
"learning_rate": 9.68316749134364e-07,
|
| 386 |
+
"loss": 0.7005,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.246458923512748,
|
| 391 |
+
"grad_norm": 0.6923178573722074,
|
| 392 |
+
"learning_rate": 9.659538714838633e-07,
|
| 393 |
+
"loss": 0.6983,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.2691218130311614,
|
| 398 |
+
"grad_norm": 0.6963394168232236,
|
| 399 |
+
"learning_rate": 9.63509121038005e-07,
|
| 400 |
+
"loss": 0.6932,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.291784702549575,
|
| 405 |
+
"grad_norm": 0.6743675615821408,
|
| 406 |
+
"learning_rate": 9.609829273641032e-07,
|
| 407 |
+
"loss": 0.6789,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.3144475920679888,
|
| 412 |
+
"grad_norm": 0.6786035246894967,
|
| 413 |
+
"learning_rate": 9.583757343398684e-07,
|
| 414 |
+
"loss": 0.6628,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.3371104815864023,
|
| 419 |
+
"grad_norm": 0.7270460673039131,
|
| 420 |
+
"learning_rate": 9.55688000075414e-07,
|
| 421 |
+
"loss": 0.6831,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.3597733711048159,
|
| 426 |
+
"grad_norm": 0.6841455902480504,
|
| 427 |
+
"learning_rate": 9.529201968327616e-07,
|
| 428 |
+
"loss": 0.6951,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.3824362606232294,
|
| 433 |
+
"grad_norm": 0.6153616879449294,
|
| 434 |
+
"learning_rate": 9.500728109428603e-07,
|
| 435 |
+
"loss": 0.676,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.405099150141643,
|
| 440 |
+
"grad_norm": 0.6177487537567523,
|
| 441 |
+
"learning_rate": 9.47146342720133e-07,
|
| 442 |
+
"loss": 0.6842,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.4277620396600565,
|
| 447 |
+
"grad_norm": 0.5753559089127149,
|
| 448 |
+
"learning_rate": 9.441413063745659e-07,
|
| 449 |
+
"loss": 0.6408,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.4504249291784703,
|
| 454 |
+
"grad_norm": 0.620464077741966,
|
| 455 |
+
"learning_rate": 9.410582299213572e-07,
|
| 456 |
+
"loss": 0.6952,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.4730878186968839,
|
| 461 |
+
"grad_norm": 0.587732312757755,
|
| 462 |
+
"learning_rate": 9.378976550881392e-07,
|
| 463 |
+
"loss": 0.6897,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.4957507082152974,
|
| 468 |
+
"grad_norm": 0.6133303288545134,
|
| 469 |
+
"learning_rate": 9.346601372197913e-07,
|
| 470 |
+
"loss": 0.6319,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.5184135977337112,
|
| 475 |
+
"grad_norm": 0.5975684805854956,
|
| 476 |
+
"learning_rate": 9.313462451808599e-07,
|
| 477 |
+
"loss": 0.7085,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.5410764872521248,
|
| 482 |
+
"grad_norm": 0.5691716789827311,
|
| 483 |
+
"learning_rate": 9.279565612556042e-07,
|
| 484 |
+
"loss": 0.6799,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.5637393767705383,
|
| 489 |
+
"grad_norm": 0.5623581760482004,
|
| 490 |
+
"learning_rate": 9.24491681045682e-07,
|
| 491 |
+
"loss": 0.6627,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.5864022662889519,
|
| 496 |
+
"grad_norm": 0.5545018113642449,
|
| 497 |
+
"learning_rate": 9.209522133654968e-07,
|
| 498 |
+
"loss": 0.6673,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.6090651558073654,
|
| 503 |
+
"grad_norm": 0.6223379664208608,
|
| 504 |
+
"learning_rate": 9.17338780135223e-07,
|
| 505 |
+
"loss": 0.6682,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.631728045325779,
|
| 510 |
+
"grad_norm": 0.5484348938274137,
|
| 511 |
+
"learning_rate": 9.136520162715286e-07,
|
| 512 |
+
"loss": 0.6459,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.6543909348441925,
|
| 517 |
+
"grad_norm": 0.598633459691356,
|
| 518 |
+
"learning_rate": 9.098925695760131e-07,
|
| 519 |
+
"loss": 0.6663,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.677053824362606,
|
| 524 |
+
"grad_norm": 0.6063642708751795,
|
| 525 |
+
"learning_rate": 9.060611006213832e-07,
|
| 526 |
+
"loss": 0.6471,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.6997167138810199,
|
| 531 |
+
"grad_norm": 0.5310843433827631,
|
| 532 |
+
"learning_rate": 9.021582826353824e-07,
|
| 533 |
+
"loss": 0.6422,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.7223796033994334,
|
| 538 |
+
"grad_norm": 0.5899701772442509,
|
| 539 |
+
"learning_rate": 8.981848013824993e-07,
|
| 540 |
+
"loss": 0.6616,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.7450424929178472,
|
| 545 |
+
"grad_norm": 0.6774981304086599,
|
| 546 |
+
"learning_rate": 8.94141355043471e-07,
|
| 547 |
+
"loss": 0.6442,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.7677053824362607,
|
| 552 |
+
"grad_norm": 0.5555862881849043,
|
| 553 |
+
"learning_rate": 8.90028654092606e-07,
|
| 554 |
+
"loss": 0.6427,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.7903682719546743,
|
| 559 |
+
"grad_norm": 0.5521769324318557,
|
| 560 |
+
"learning_rate": 8.858474211729469e-07,
|
| 561 |
+
"loss": 0.6308,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.8130311614730878,
|
| 566 |
+
"grad_norm": 0.5094008328024741,
|
| 567 |
+
"learning_rate": 8.815983909692941e-07,
|
| 568 |
+
"loss": 0.6375,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.8356940509915014,
|
| 573 |
+
"grad_norm": 0.47949684902186096,
|
| 574 |
+
"learning_rate": 8.77282310079115e-07,
|
| 575 |
+
"loss": 0.6124,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.858356940509915,
|
| 580 |
+
"grad_norm": 0.5457213358478963,
|
| 581 |
+
"learning_rate": 8.72899936881359e-07,
|
| 582 |
+
"loss": 0.676,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.8810198300283285,
|
| 587 |
+
"grad_norm": 0.5475114660934921,
|
| 588 |
+
"learning_rate": 8.684520414032023e-07,
|
| 589 |
+
"loss": 0.6462,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.903682719546742,
|
| 594 |
+
"grad_norm": 0.5780771596548755,
|
| 595 |
+
"learning_rate": 8.639394051847471e-07,
|
| 596 |
+
"loss": 0.629,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.9263456090651558,
|
| 601 |
+
"grad_norm": 0.5153044368837152,
|
| 602 |
+
"learning_rate": 8.593628211416963e-07,
|
| 603 |
+
"loss": 0.6607,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.9490084985835694,
|
| 608 |
+
"grad_norm": 0.5078347714748787,
|
| 609 |
+
"learning_rate": 8.547230934260311e-07,
|
| 610 |
+
"loss": 0.653,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.9716713881019832,
|
| 615 |
+
"grad_norm": 0.5090369208403657,
|
| 616 |
+
"learning_rate": 8.500210372847126e-07,
|
| 617 |
+
"loss": 0.6555,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.9943342776203967,
|
| 622 |
+
"grad_norm": 0.521639825746896,
|
| 623 |
+
"learning_rate": 8.45257478916435e-07,
|
| 624 |
+
"loss": 0.6187,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.0169971671388103,
|
| 629 |
+
"grad_norm": 1.4410319682327064,
|
| 630 |
+
"learning_rate": 8.404332553264546e-07,
|
| 631 |
+
"loss": 1.1825,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.039660056657224,
|
| 636 |
+
"grad_norm": 0.5362038209234066,
|
| 637 |
+
"learning_rate": 8.355492141795184e-07,
|
| 638 |
+
"loss": 0.6046,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.0623229461756374,
|
| 643 |
+
"grad_norm": 0.5295224430525873,
|
| 644 |
+
"learning_rate": 8.306062136509219e-07,
|
| 645 |
+
"loss": 0.607,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.084985835694051,
|
| 650 |
+
"grad_norm": 0.5438204610049183,
|
| 651 |
+
"learning_rate": 8.256051222757187e-07,
|
| 652 |
+
"loss": 0.6425,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.1076487252124645,
|
| 657 |
+
"grad_norm": 0.5637438849056178,
|
| 658 |
+
"learning_rate": 8.2054681879611e-07,
|
| 659 |
+
"loss": 0.6472,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.130311614730878,
|
| 664 |
+
"grad_norm": 0.4915626737833171,
|
| 665 |
+
"learning_rate": 8.154321920070412e-07,
|
| 666 |
+
"loss": 0.6366,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.1529745042492916,
|
| 671 |
+
"grad_norm": 0.5445000714826581,
|
| 672 |
+
"learning_rate": 8.102621406000308e-07,
|
| 673 |
+
"loss": 0.6302,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.1756373937677056,
|
| 678 |
+
"grad_norm": 0.544017574639994,
|
| 679 |
+
"learning_rate": 8.050375730052621e-07,
|
| 680 |
+
"loss": 0.6016,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.198300283286119,
|
| 685 |
+
"grad_norm": 0.7667138278664033,
|
| 686 |
+
"learning_rate": 7.997594072319625e-07,
|
| 687 |
+
"loss": 0.6476,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.2209631728045327,
|
| 692 |
+
"grad_norm": 0.5723261101431134,
|
| 693 |
+
"learning_rate": 7.944285707070997e-07,
|
| 694 |
+
"loss": 0.5982,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.2436260623229463,
|
| 699 |
+
"grad_norm": 0.5198427284810859,
|
| 700 |
+
"learning_rate": 7.890460001124241e-07,
|
| 701 |
+
"loss": 0.6373,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.26628895184136,
|
| 706 |
+
"grad_norm": 0.5082652201383684,
|
| 707 |
+
"learning_rate": 7.83612641219884e-07,
|
| 708 |
+
"loss": 0.5894,
|
| 709 |
+
"step": 100
|
| 710 |
+
}
|
| 711 |
+
],
|
| 712 |
+
"logging_steps": 1,
|
| 713 |
+
"max_steps": 264,
|
| 714 |
+
"num_input_tokens_seen": 0,
|
| 715 |
+
"num_train_epochs": 6,
|
| 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": 208143843852288.0,
|
| 730 |
+
"train_batch_size": 4,
|
| 731 |
+
"trial_name": null,
|
| 732 |
+
"trial_params": null
|
| 733 |
+
}
|