Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +716 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,716 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 468,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004273504273504274,
|
| 13 |
+
"grad_norm": 20.463779905967794,
|
| 14 |
+
"learning_rate": 4.2553191489361704e-07,
|
| 15 |
+
"loss": 1.8792,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.021367521367521368,
|
| 20 |
+
"grad_norm": 18.25002365272532,
|
| 21 |
+
"learning_rate": 2.1276595744680853e-06,
|
| 22 |
+
"loss": 1.8169,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.042735042735042736,
|
| 27 |
+
"grad_norm": 5.520763657053264,
|
| 28 |
+
"learning_rate": 4.255319148936171e-06,
|
| 29 |
+
"loss": 1.5621,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0641025641025641,
|
| 34 |
+
"grad_norm": 2.4039272335616273,
|
| 35 |
+
"learning_rate": 6.382978723404256e-06,
|
| 36 |
+
"loss": 1.4414,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08547008547008547,
|
| 41 |
+
"grad_norm": 1.5018627664657522,
|
| 42 |
+
"learning_rate": 8.510638297872341e-06,
|
| 43 |
+
"loss": 1.3849,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.10683760683760683,
|
| 48 |
+
"grad_norm": 1.2879097291220103,
|
| 49 |
+
"learning_rate": 1.0638297872340426e-05,
|
| 50 |
+
"loss": 1.3222,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1282051282051282,
|
| 55 |
+
"grad_norm": 1.2033621208548797,
|
| 56 |
+
"learning_rate": 1.2765957446808513e-05,
|
| 57 |
+
"loss": 1.3365,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14957264957264957,
|
| 62 |
+
"grad_norm": 1.1255514254267025,
|
| 63 |
+
"learning_rate": 1.4893617021276596e-05,
|
| 64 |
+
"loss": 1.2732,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.17094017094017094,
|
| 69 |
+
"grad_norm": 1.243442983004785,
|
| 70 |
+
"learning_rate": 1.7021276595744682e-05,
|
| 71 |
+
"loss": 1.2952,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.19230769230769232,
|
| 76 |
+
"grad_norm": 1.398459861930405,
|
| 77 |
+
"learning_rate": 1.914893617021277e-05,
|
| 78 |
+
"loss": 1.2762,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.21367521367521367,
|
| 83 |
+
"grad_norm": 1.871011686158811,
|
| 84 |
+
"learning_rate": 1.999749429505675e-05,
|
| 85 |
+
"loss": 1.2997,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.23504273504273504,
|
| 90 |
+
"grad_norm": 1.2331867317142813,
|
| 91 |
+
"learning_rate": 1.9982186200932964e-05,
|
| 92 |
+
"loss": 1.2568,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2564102564102564,
|
| 97 |
+
"grad_norm": 1.1946705653350014,
|
| 98 |
+
"learning_rate": 1.9952983353325358e-05,
|
| 99 |
+
"loss": 1.2763,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2777777777777778,
|
| 104 |
+
"grad_norm": 1.1214723834130367,
|
| 105 |
+
"learning_rate": 1.990992640128218e-05,
|
| 106 |
+
"loss": 1.2938,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.29914529914529914,
|
| 111 |
+
"grad_norm": 1.119663106979562,
|
| 112 |
+
"learning_rate": 1.9853075278140913e-05,
|
| 113 |
+
"loss": 1.2598,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.32051282051282054,
|
| 118 |
+
"grad_norm": 1.1303932474940224,
|
| 119 |
+
"learning_rate": 1.9782509118103773e-05,
|
| 120 |
+
"loss": 1.2489,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3418803418803419,
|
| 125 |
+
"grad_norm": 1.2099240945757408,
|
| 126 |
+
"learning_rate": 1.9698326146086446e-05,
|
| 127 |
+
"loss": 1.2813,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.36324786324786323,
|
| 132 |
+
"grad_norm": 1.108377660801764,
|
| 133 |
+
"learning_rate": 1.9600643540993453e-05,
|
| 134 |
+
"loss": 1.2296,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.38461538461538464,
|
| 139 |
+
"grad_norm": 1.1652590169175916,
|
| 140 |
+
"learning_rate": 1.9489597272610377e-05,
|
| 141 |
+
"loss": 1.2367,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.405982905982906,
|
| 146 |
+
"grad_norm": 1.144744441272463,
|
| 147 |
+
"learning_rate": 1.936534191234006e-05,
|
| 148 |
+
"loss": 1.2497,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.42735042735042733,
|
| 153 |
+
"grad_norm": 1.0797546591912566,
|
| 154 |
+
"learning_rate": 1.922805041804617e-05,
|
| 155 |
+
"loss": 1.2157,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.44871794871794873,
|
| 160 |
+
"grad_norm": 1.1997371054541615,
|
| 161 |
+
"learning_rate": 1.907791389330363e-05,
|
| 162 |
+
"loss": 1.2393,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4700854700854701,
|
| 167 |
+
"grad_norm": 1.234499931127894,
|
| 168 |
+
"learning_rate": 1.8915141321391083e-05,
|
| 169 |
+
"loss": 1.268,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.49145299145299143,
|
| 174 |
+
"grad_norm": 1.2719307847654406,
|
| 175 |
+
"learning_rate": 1.873995927439555e-05,
|
| 176 |
+
"loss": 1.2318,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5128205128205128,
|
| 181 |
+
"grad_norm": 1.0803159349159561,
|
| 182 |
+
"learning_rate": 1.855261159783432e-05,
|
| 183 |
+
"loss": 1.2422,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5341880341880342,
|
| 188 |
+
"grad_norm": 1.1126726338850874,
|
| 189 |
+
"learning_rate": 1.8353359071232954e-05,
|
| 190 |
+
"loss": 1.2341,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5555555555555556,
|
| 195 |
+
"grad_norm": 1.141289352623837,
|
| 196 |
+
"learning_rate": 1.8142479045131956e-05,
|
| 197 |
+
"loss": 1.2172,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5769230769230769,
|
| 202 |
+
"grad_norm": 1.0891200836049277,
|
| 203 |
+
"learning_rate": 1.7920265055027285e-05,
|
| 204 |
+
"loss": 1.2147,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5982905982905983,
|
| 209 |
+
"grad_norm": 0.9983723250381306,
|
| 210 |
+
"learning_rate": 1.76870264127822e-05,
|
| 211 |
+
"loss": 1.2087,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6196581196581197,
|
| 216 |
+
"grad_norm": 1.0642854505735857,
|
| 217 |
+
"learning_rate": 1.7443087776079068e-05,
|
| 218 |
+
"loss": 1.1872,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6410256410256411,
|
| 223 |
+
"grad_norm": 1.0890040938005288,
|
| 224 |
+
"learning_rate": 1.7188788696510477e-05,
|
| 225 |
+
"loss": 1.2431,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6623931623931624,
|
| 230 |
+
"grad_norm": 1.0196805310002042,
|
| 231 |
+
"learning_rate": 1.6924483146938756e-05,
|
| 232 |
+
"loss": 1.2133,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6837606837606838,
|
| 237 |
+
"grad_norm": 1.0422167588733457,
|
| 238 |
+
"learning_rate": 1.665053902878167e-05,
|
| 239 |
+
"loss": 1.2275,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7051282051282052,
|
| 244 |
+
"grad_norm": 1.0099548152705635,
|
| 245 |
+
"learning_rate": 1.6367337659910223e-05,
|
| 246 |
+
"loss": 1.1997,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7264957264957265,
|
| 251 |
+
"grad_norm": 1.034450507226083,
|
| 252 |
+
"learning_rate": 1.607527324387137e-05,
|
| 253 |
+
"loss": 1.1952,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7478632478632479,
|
| 258 |
+
"grad_norm": 1.01064689098683,
|
| 259 |
+
"learning_rate": 1.5774752321174428e-05,
|
| 260 |
+
"loss": 1.2142,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7692307692307693,
|
| 265 |
+
"grad_norm": 1.2440710808413185,
|
| 266 |
+
"learning_rate": 1.5466193203405017e-05,
|
| 267 |
+
"loss": 1.2272,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.7905982905982906,
|
| 272 |
+
"grad_norm": 0.9538688514209636,
|
| 273 |
+
"learning_rate": 1.5150025390954153e-05,
|
| 274 |
+
"loss": 1.1797,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.811965811965812,
|
| 279 |
+
"grad_norm": 1.0457627221644799,
|
| 280 |
+
"learning_rate": 1.4826688975173085e-05,
|
| 281 |
+
"loss": 1.1834,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8333333333333334,
|
| 286 |
+
"grad_norm": 1.0494277738904705,
|
| 287 |
+
"learning_rate": 1.4496634025785938e-05,
|
| 288 |
+
"loss": 1.2098,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8547008547008547,
|
| 293 |
+
"grad_norm": 0.979634991944023,
|
| 294 |
+
"learning_rate": 1.4160319964412943e-05,
|
| 295 |
+
"loss": 1.1936,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8760683760683761,
|
| 300 |
+
"grad_norm": 1.0707731906559956,
|
| 301 |
+
"learning_rate": 1.3818214925076226e-05,
|
| 302 |
+
"loss": 1.1934,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.8974358974358975,
|
| 307 |
+
"grad_norm": 0.9902950125875686,
|
| 308 |
+
"learning_rate": 1.3470795102578358e-05,
|
| 309 |
+
"loss": 1.1875,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9188034188034188,
|
| 314 |
+
"grad_norm": 1.1906219730782999,
|
| 315 |
+
"learning_rate": 1.3118544089660635e-05,
|
| 316 |
+
"loss": 1.1828,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9401709401709402,
|
| 321 |
+
"grad_norm": 1.029497385097088,
|
| 322 |
+
"learning_rate": 1.2761952203863759e-05,
|
| 323 |
+
"loss": 1.2032,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.9615384615384616,
|
| 328 |
+
"grad_norm": 1.0145349544318898,
|
| 329 |
+
"learning_rate": 1.2401515805027924e-05,
|
| 330 |
+
"loss": 1.1918,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.9829059829059829,
|
| 335 |
+
"grad_norm": 0.9338647888184574,
|
| 336 |
+
"learning_rate": 1.2037736604382279e-05,
|
| 337 |
+
"loss": 1.1667,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.0,
|
| 342 |
+
"eval_loss": 1.1897046566009521,
|
| 343 |
+
"eval_runtime": 4.2537,
|
| 344 |
+
"eval_samples_per_second": 34.323,
|
| 345 |
+
"eval_steps_per_second": 0.705,
|
| 346 |
+
"step": 234
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.0042735042735043,
|
| 350 |
+
"grad_norm": 2.061617781988495,
|
| 351 |
+
"learning_rate": 1.1671120966185486e-05,
|
| 352 |
+
"loss": 1.1272,
|
| 353 |
+
"step": 235
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.0256410256410255,
|
| 357 |
+
"grad_norm": 1.5757999937786025,
|
| 358 |
+
"learning_rate": 1.1302179202889505e-05,
|
| 359 |
+
"loss": 0.8684,
|
| 360 |
+
"step": 240
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.047008547008547,
|
| 364 |
+
"grad_norm": 1.162411965496565,
|
| 365 |
+
"learning_rate": 1.0931424864807624e-05,
|
| 366 |
+
"loss": 0.8433,
|
| 367 |
+
"step": 245
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.0683760683760684,
|
| 371 |
+
"grad_norm": 1.083844459735895,
|
| 372 |
+
"learning_rate": 1.0559374025275597e-05,
|
| 373 |
+
"loss": 0.8588,
|
| 374 |
+
"step": 250
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.0897435897435896,
|
| 378 |
+
"grad_norm": 1.0166749425497479,
|
| 379 |
+
"learning_rate": 1.0186544562300766e-05,
|
| 380 |
+
"loss": 0.8936,
|
| 381 |
+
"step": 255
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.1111111111111112,
|
| 385 |
+
"grad_norm": 1.0711811120340167,
|
| 386 |
+
"learning_rate": 9.813455437699238e-06,
|
| 387 |
+
"loss": 0.8633,
|
| 388 |
+
"step": 260
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.1324786324786325,
|
| 392 |
+
"grad_norm": 1.0038627368222364,
|
| 393 |
+
"learning_rate": 9.440625974724408e-06,
|
| 394 |
+
"loss": 0.8821,
|
| 395 |
+
"step": 265
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.1538461538461537,
|
| 399 |
+
"grad_norm": 1.0431997146297058,
|
| 400 |
+
"learning_rate": 9.068575135192377e-06,
|
| 401 |
+
"loss": 0.8856,
|
| 402 |
+
"step": 270
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.1752136752136753,
|
| 406 |
+
"grad_norm": 1.4331559261347666,
|
| 407 |
+
"learning_rate": 8.697820797110499e-06,
|
| 408 |
+
"loss": 0.8645,
|
| 409 |
+
"step": 275
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.1965811965811965,
|
| 413 |
+
"grad_norm": 1.0197402946112883,
|
| 414 |
+
"learning_rate": 8.328879033814516e-06,
|
| 415 |
+
"loss": 0.8512,
|
| 416 |
+
"step": 280
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.217948717948718,
|
| 420 |
+
"grad_norm": 1.01821032251223,
|
| 421 |
+
"learning_rate": 7.962263395617724e-06,
|
| 422 |
+
"loss": 0.8661,
|
| 423 |
+
"step": 285
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.2393162393162394,
|
| 427 |
+
"grad_norm": 1.0267748605334936,
|
| 428 |
+
"learning_rate": 7.598484194972076e-06,
|
| 429 |
+
"loss": 0.8601,
|
| 430 |
+
"step": 290
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.2606837606837606,
|
| 434 |
+
"grad_norm": 1.0208240145129162,
|
| 435 |
+
"learning_rate": 7.238047796136247e-06,
|
| 436 |
+
"loss": 0.8627,
|
| 437 |
+
"step": 295
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.282051282051282,
|
| 441 |
+
"grad_norm": 1.003007647188531,
|
| 442 |
+
"learning_rate": 6.881455910339369e-06,
|
| 443 |
+
"loss": 0.8463,
|
| 444 |
+
"step": 300
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.3034188034188035,
|
| 448 |
+
"grad_norm": 1.0588099942878915,
|
| 449 |
+
"learning_rate": 6.529204897421644e-06,
|
| 450 |
+
"loss": 0.8591,
|
| 451 |
+
"step": 305
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.3247863247863247,
|
| 455 |
+
"grad_norm": 0.9819416599015215,
|
| 456 |
+
"learning_rate": 6.181785074923778e-06,
|
| 457 |
+
"loss": 0.863,
|
| 458 |
+
"step": 310
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.3461538461538463,
|
| 462 |
+
"grad_norm": 0.9892062975951394,
|
| 463 |
+
"learning_rate": 5.839680035587061e-06,
|
| 464 |
+
"loss": 0.8594,
|
| 465 |
+
"step": 315
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.3675213675213675,
|
| 469 |
+
"grad_norm": 0.9795643152120956,
|
| 470 |
+
"learning_rate": 5.503365974214059e-06,
|
| 471 |
+
"loss": 0.8613,
|
| 472 |
+
"step": 320
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.3888888888888888,
|
| 476 |
+
"grad_norm": 0.9542734758696676,
|
| 477 |
+
"learning_rate": 5.173311024826916e-06,
|
| 478 |
+
"loss": 0.866,
|
| 479 |
+
"step": 325
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.4102564102564101,
|
| 483 |
+
"grad_norm": 0.9352942170714407,
|
| 484 |
+
"learning_rate": 4.849974609045849e-06,
|
| 485 |
+
"loss": 0.8469,
|
| 486 |
+
"step": 330
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.4316239316239316,
|
| 490 |
+
"grad_norm": 0.943859757278146,
|
| 491 |
+
"learning_rate": 4.533806796594989e-06,
|
| 492 |
+
"loss": 0.8729,
|
| 493 |
+
"step": 335
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.452991452991453,
|
| 497 |
+
"grad_norm": 1.058117869970705,
|
| 498 |
+
"learning_rate": 4.2252476788255735e-06,
|
| 499 |
+
"loss": 0.8614,
|
| 500 |
+
"step": 340
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.4743589743589745,
|
| 504 |
+
"grad_norm": 1.007136340214763,
|
| 505 |
+
"learning_rate": 3.924726756128632e-06,
|
| 506 |
+
"loss": 0.8314,
|
| 507 |
+
"step": 345
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.4957264957264957,
|
| 511 |
+
"grad_norm": 1.0001072453743785,
|
| 512 |
+
"learning_rate": 3.6326623400897797e-06,
|
| 513 |
+
"loss": 0.8263,
|
| 514 |
+
"step": 350
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.517094017094017,
|
| 518 |
+
"grad_norm": 11.67014141174618,
|
| 519 |
+
"learning_rate": 3.3494609712183323e-06,
|
| 520 |
+
"loss": 0.9014,
|
| 521 |
+
"step": 355
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.5384615384615383,
|
| 525 |
+
"grad_norm": 0.9396806102918359,
|
| 526 |
+
"learning_rate": 3.0755168530612444e-06,
|
| 527 |
+
"loss": 0.8294,
|
| 528 |
+
"step": 360
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.5598290598290598,
|
| 532 |
+
"grad_norm": 0.9899965091834676,
|
| 533 |
+
"learning_rate": 2.8112113034895273e-06,
|
| 534 |
+
"loss": 0.8451,
|
| 535 |
+
"step": 365
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.5811965811965814,
|
| 539 |
+
"grad_norm": 1.0046619245072979,
|
| 540 |
+
"learning_rate": 2.5569122239209366e-06,
|
| 541 |
+
"loss": 0.8635,
|
| 542 |
+
"step": 370
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.6025641025641026,
|
| 546 |
+
"grad_norm": 0.9398590773976044,
|
| 547 |
+
"learning_rate": 2.312973587217798e-06,
|
| 548 |
+
"loss": 0.8468,
|
| 549 |
+
"step": 375
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.623931623931624,
|
| 553 |
+
"grad_norm": 0.9897283589814032,
|
| 554 |
+
"learning_rate": 2.079734944972717e-06,
|
| 555 |
+
"loss": 0.8615,
|
| 556 |
+
"step": 380
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.6452991452991452,
|
| 560 |
+
"grad_norm": 0.9311555025619238,
|
| 561 |
+
"learning_rate": 1.8575209548680472e-06,
|
| 562 |
+
"loss": 0.8616,
|
| 563 |
+
"step": 385
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.6666666666666665,
|
| 567 |
+
"grad_norm": 0.9902450799068546,
|
| 568 |
+
"learning_rate": 1.646640928767047e-06,
|
| 569 |
+
"loss": 0.8401,
|
| 570 |
+
"step": 390
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.688034188034188,
|
| 574 |
+
"grad_norm": 0.9397904063208168,
|
| 575 |
+
"learning_rate": 1.447388402165686e-06,
|
| 576 |
+
"loss": 0.8287,
|
| 577 |
+
"step": 395
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.7094017094017095,
|
| 581 |
+
"grad_norm": 0.9334825682417511,
|
| 582 |
+
"learning_rate": 1.2600407256044544e-06,
|
| 583 |
+
"loss": 0.8492,
|
| 584 |
+
"step": 400
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.7307692307692308,
|
| 588 |
+
"grad_norm": 0.9901247743535753,
|
| 589 |
+
"learning_rate": 1.084858678608922e-06,
|
| 590 |
+
"loss": 0.8695,
|
| 591 |
+
"step": 405
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.7521367521367521,
|
| 595 |
+
"grad_norm": 0.9223418850643341,
|
| 596 |
+
"learning_rate": 9.220861066963715e-07,
|
| 597 |
+
"loss": 0.8476,
|
| 598 |
+
"step": 410
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.7735042735042734,
|
| 602 |
+
"grad_norm": 0.9421948230776865,
|
| 603 |
+
"learning_rate": 7.719495819538325e-07,
|
| 604 |
+
"loss": 0.8412,
|
| 605 |
+
"step": 415
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.7948717948717947,
|
| 609 |
+
"grad_norm": 0.9548113796103942,
|
| 610 |
+
"learning_rate": 6.346580876599395e-07,
|
| 611 |
+
"loss": 0.8633,
|
| 612 |
+
"step": 420
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.8162393162393162,
|
| 616 |
+
"grad_norm": 0.9286412348527092,
|
| 617 |
+
"learning_rate": 5.104027273896239e-07,
|
| 618 |
+
"loss": 0.8659,
|
| 619 |
+
"step": 425
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.8376068376068377,
|
| 623 |
+
"grad_norm": 0.9726444202093633,
|
| 624 |
+
"learning_rate": 3.9935645900654906e-07,
|
| 625 |
+
"loss": 0.8533,
|
| 626 |
+
"step": 430
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.858974358974359,
|
| 630 |
+
"grad_norm": 0.9864530246870579,
|
| 631 |
+
"learning_rate": 3.016738539135566e-07,
|
| 632 |
+
"loss": 0.8377,
|
| 633 |
+
"step": 435
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.8803418803418803,
|
| 637 |
+
"grad_norm": 0.9665706648338857,
|
| 638 |
+
"learning_rate": 2.1749088189622848e-07,
|
| 639 |
+
"loss": 0.8411,
|
| 640 |
+
"step": 440
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.9017094017094016,
|
| 644 |
+
"grad_norm": 0.8848822721798258,
|
| 645 |
+
"learning_rate": 1.4692472185908635e-07,
|
| 646 |
+
"loss": 0.8615,
|
| 647 |
+
"step": 445
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.9230769230769231,
|
| 651 |
+
"grad_norm": 0.9643097064713656,
|
| 652 |
+
"learning_rate": 9.00735987178214e-08,
|
| 653 |
+
"loss": 0.8367,
|
| 654 |
+
"step": 450
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.9444444444444444,
|
| 658 |
+
"grad_norm": 0.9452651983857008,
|
| 659 |
+
"learning_rate": 4.701664667464245e-08,
|
| 660 |
+
"loss": 0.8577,
|
| 661 |
+
"step": 455
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.965811965811966,
|
| 665 |
+
"grad_norm": 0.9405976480323978,
|
| 666 |
+
"learning_rate": 1.781379906703573e-08,
|
| 667 |
+
"loss": 0.8518,
|
| 668 |
+
"step": 460
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.9871794871794872,
|
| 672 |
+
"grad_norm": 0.9339891238206331,
|
| 673 |
+
"learning_rate": 2.5057049432519744e-09,
|
| 674 |
+
"loss": 0.8465,
|
| 675 |
+
"step": 465
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.0,
|
| 679 |
+
"eval_loss": 1.2036917209625244,
|
| 680 |
+
"eval_runtime": 4.1091,
|
| 681 |
+
"eval_samples_per_second": 35.531,
|
| 682 |
+
"eval_steps_per_second": 0.73,
|
| 683 |
+
"step": 468
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.0,
|
| 687 |
+
"step": 468,
|
| 688 |
+
"total_flos": 97989678858240.0,
|
| 689 |
+
"train_loss": 1.058635287050508,
|
| 690 |
+
"train_runtime": 3412.8477,
|
| 691 |
+
"train_samples_per_second": 8.744,
|
| 692 |
+
"train_steps_per_second": 0.137
|
| 693 |
+
}
|
| 694 |
+
],
|
| 695 |
+
"logging_steps": 5,
|
| 696 |
+
"max_steps": 468,
|
| 697 |
+
"num_input_tokens_seen": 0,
|
| 698 |
+
"num_train_epochs": 2,
|
| 699 |
+
"save_steps": 500,
|
| 700 |
+
"stateful_callbacks": {
|
| 701 |
+
"TrainerControl": {
|
| 702 |
+
"args": {
|
| 703 |
+
"should_epoch_stop": false,
|
| 704 |
+
"should_evaluate": false,
|
| 705 |
+
"should_log": false,
|
| 706 |
+
"should_save": true,
|
| 707 |
+
"should_training_stop": true
|
| 708 |
+
},
|
| 709 |
+
"attributes": {}
|
| 710 |
+
}
|
| 711 |
+
},
|
| 712 |
+
"total_flos": 97989678858240.0,
|
| 713 |
+
"train_batch_size": 8,
|
| 714 |
+
"trial_name": null,
|
| 715 |
+
"trial_params": null
|
| 716 |
+
}
|