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": 0.032797638570022956,
|
| 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": 0.00032797638570022957,
|
| 13 |
+
"grad_norm": 3.994058609008789,
|
| 14 |
+
"learning_rate": 0.00019805941782534764,
|
| 15 |
+
"loss": 2.1587,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0006559527714004591,
|
| 20 |
+
"grad_norm": 3.933835983276367,
|
| 21 |
+
"learning_rate": 0.00019605881764529358,
|
| 22 |
+
"loss": 2.0517,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0009839291571006887,
|
| 27 |
+
"grad_norm": 2.435616970062256,
|
| 28 |
+
"learning_rate": 0.00019405821746523957,
|
| 29 |
+
"loss": 1.6092,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0013119055428009183,
|
| 34 |
+
"grad_norm": 2.745454788208008,
|
| 35 |
+
"learning_rate": 0.00019205761728518557,
|
| 36 |
+
"loss": 1.4157,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0016398819285011479,
|
| 41 |
+
"grad_norm": 1.7035468816757202,
|
| 42 |
+
"learning_rate": 0.00019005701710513156,
|
| 43 |
+
"loss": 1.0786,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0019678583142013774,
|
| 48 |
+
"grad_norm": 2.8651487827301025,
|
| 49 |
+
"learning_rate": 0.00018805641692507753,
|
| 50 |
+
"loss": 1.0119,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0022958346999016072,
|
| 55 |
+
"grad_norm": 3.744490146636963,
|
| 56 |
+
"learning_rate": 0.00018605581674502352,
|
| 57 |
+
"loss": 0.9555,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0026238110856018366,
|
| 62 |
+
"grad_norm": 1.1622904539108276,
|
| 63 |
+
"learning_rate": 0.00018405521656496952,
|
| 64 |
+
"loss": 0.8364,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0029517874713020664,
|
| 69 |
+
"grad_norm": 8.87141227722168,
|
| 70 |
+
"learning_rate": 0.00018205461638491548,
|
| 71 |
+
"loss": 0.7568,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0032797638570022957,
|
| 76 |
+
"grad_norm": 1.4474456310272217,
|
| 77 |
+
"learning_rate": 0.00018005401620486148,
|
| 78 |
+
"loss": 0.7653,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0036077402427025255,
|
| 83 |
+
"grad_norm": 1.3203747272491455,
|
| 84 |
+
"learning_rate": 0.00017805341602480744,
|
| 85 |
+
"loss": 0.643,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.003935716628402755,
|
| 90 |
+
"grad_norm": 1.6685700416564941,
|
| 91 |
+
"learning_rate": 0.00017605281584475344,
|
| 92 |
+
"loss": 0.6993,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.004263693014102984,
|
| 97 |
+
"grad_norm": 1.86400306224823,
|
| 98 |
+
"learning_rate": 0.00017405221566469943,
|
| 99 |
+
"loss": 0.6262,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0045916693998032145,
|
| 104 |
+
"grad_norm": 1.2344428300857544,
|
| 105 |
+
"learning_rate": 0.00017205161548464542,
|
| 106 |
+
"loss": 0.6256,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.004919645785503444,
|
| 111 |
+
"grad_norm": 1.351318359375,
|
| 112 |
+
"learning_rate": 0.00017005101530459136,
|
| 113 |
+
"loss": 0.6516,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.005247622171203673,
|
| 118 |
+
"grad_norm": 1.4530205726623535,
|
| 119 |
+
"learning_rate": 0.00016805041512453736,
|
| 120 |
+
"loss": 0.6692,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0055755985569039025,
|
| 125 |
+
"grad_norm": 1.2419607639312744,
|
| 126 |
+
"learning_rate": 0.00016604981494448335,
|
| 127 |
+
"loss": 0.6294,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.005903574942604133,
|
| 132 |
+
"grad_norm": 1.0507097244262695,
|
| 133 |
+
"learning_rate": 0.00016404921476442935,
|
| 134 |
+
"loss": 0.5902,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.006231551328304362,
|
| 139 |
+
"grad_norm": 0.8393335938453674,
|
| 140 |
+
"learning_rate": 0.0001620486145843753,
|
| 141 |
+
"loss": 0.604,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0065595277140045915,
|
| 146 |
+
"grad_norm": 0.9332450032234192,
|
| 147 |
+
"learning_rate": 0.0001600480144043213,
|
| 148 |
+
"loss": 0.5327,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.006887504099704822,
|
| 153 |
+
"grad_norm": 1.0808994770050049,
|
| 154 |
+
"learning_rate": 0.0001580474142242673,
|
| 155 |
+
"loss": 0.596,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.007215480485405051,
|
| 160 |
+
"grad_norm": 0.8101345896720886,
|
| 161 |
+
"learning_rate": 0.00015604681404421327,
|
| 162 |
+
"loss": 0.5003,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.00754345687110528,
|
| 167 |
+
"grad_norm": 1.1937756538391113,
|
| 168 |
+
"learning_rate": 0.00015404621386415926,
|
| 169 |
+
"loss": 0.5872,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.00787143325680551,
|
| 174 |
+
"grad_norm": 0.9010137915611267,
|
| 175 |
+
"learning_rate": 0.00015204561368410523,
|
| 176 |
+
"loss": 0.5776,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.008199409642505739,
|
| 181 |
+
"grad_norm": 0.7353044152259827,
|
| 182 |
+
"learning_rate": 0.00015004501350405122,
|
| 183 |
+
"loss": 0.5487,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.008527386028205968,
|
| 188 |
+
"grad_norm": 1.011644721031189,
|
| 189 |
+
"learning_rate": 0.00014804441332399721,
|
| 190 |
+
"loss": 0.583,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.0088553624139062,
|
| 195 |
+
"grad_norm": 0.9079630970954895,
|
| 196 |
+
"learning_rate": 0.0001460438131439432,
|
| 197 |
+
"loss": 0.5559,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.009183338799606429,
|
| 202 |
+
"grad_norm": 0.607328474521637,
|
| 203 |
+
"learning_rate": 0.00014404321296388918,
|
| 204 |
+
"loss": 0.6093,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.009511315185306658,
|
| 209 |
+
"grad_norm": 0.8753256797790527,
|
| 210 |
+
"learning_rate": 0.00014204261278383514,
|
| 211 |
+
"loss": 0.5146,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.009839291571006888,
|
| 216 |
+
"grad_norm": 0.7783370614051819,
|
| 217 |
+
"learning_rate": 0.00014004201260378114,
|
| 218 |
+
"loss": 0.4971,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.010167267956707117,
|
| 223 |
+
"grad_norm": 0.5913918018341064,
|
| 224 |
+
"learning_rate": 0.00013804141242372713,
|
| 225 |
+
"loss": 0.5077,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.010495244342407346,
|
| 230 |
+
"grad_norm": 0.817452609539032,
|
| 231 |
+
"learning_rate": 0.00013604081224367312,
|
| 232 |
+
"loss": 0.5969,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.010823220728107576,
|
| 237 |
+
"grad_norm": 0.8791104555130005,
|
| 238 |
+
"learning_rate": 0.0001340402120636191,
|
| 239 |
+
"loss": 0.5425,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.011151197113807805,
|
| 244 |
+
"grad_norm": 0.6707533001899719,
|
| 245 |
+
"learning_rate": 0.00013203961188356508,
|
| 246 |
+
"loss": 0.5647,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.011479173499508036,
|
| 251 |
+
"grad_norm": 0.8518595099449158,
|
| 252 |
+
"learning_rate": 0.00013003901170351108,
|
| 253 |
+
"loss": 0.529,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.011807149885208265,
|
| 258 |
+
"grad_norm": 0.8530653715133667,
|
| 259 |
+
"learning_rate": 0.00012803841152345704,
|
| 260 |
+
"loss": 0.4994,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.012135126270908495,
|
| 265 |
+
"grad_norm": 0.5490475296974182,
|
| 266 |
+
"learning_rate": 0.000126037811343403,
|
| 267 |
+
"loss": 0.5335,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.012463102656608724,
|
| 272 |
+
"grad_norm": 0.6268607974052429,
|
| 273 |
+
"learning_rate": 0.000124037211163349,
|
| 274 |
+
"loss": 0.5691,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.012791079042308954,
|
| 279 |
+
"grad_norm": 0.7266058325767517,
|
| 280 |
+
"learning_rate": 0.000122036610983295,
|
| 281 |
+
"loss": 0.5389,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.013119055428009183,
|
| 286 |
+
"grad_norm": 0.6403557062149048,
|
| 287 |
+
"learning_rate": 0.00012003601080324098,
|
| 288 |
+
"loss": 0.5405,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.013447031813709412,
|
| 293 |
+
"grad_norm": 0.5774125456809998,
|
| 294 |
+
"learning_rate": 0.00011803541062318697,
|
| 295 |
+
"loss": 0.5031,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.013775008199409643,
|
| 300 |
+
"grad_norm": 0.7147932052612305,
|
| 301 |
+
"learning_rate": 0.00011603481044313295,
|
| 302 |
+
"loss": 0.5353,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.014102984585109873,
|
| 307 |
+
"grad_norm": 0.5923655033111572,
|
| 308 |
+
"learning_rate": 0.00011403421026307892,
|
| 309 |
+
"loss": 0.5532,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.014430960970810102,
|
| 314 |
+
"grad_norm": 0.6512230634689331,
|
| 315 |
+
"learning_rate": 0.00011203361008302491,
|
| 316 |
+
"loss": 0.5585,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.014758937356510331,
|
| 321 |
+
"grad_norm": 0.5685781240463257,
|
| 322 |
+
"learning_rate": 0.0001100330099029709,
|
| 323 |
+
"loss": 0.5231,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.01508691374221056,
|
| 328 |
+
"grad_norm": 0.5105693936347961,
|
| 329 |
+
"learning_rate": 0.00010803240972291689,
|
| 330 |
+
"loss": 0.5082,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.01541489012791079,
|
| 335 |
+
"grad_norm": 0.5665640234947205,
|
| 336 |
+
"learning_rate": 0.00010603180954286287,
|
| 337 |
+
"loss": 0.551,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.01574286651361102,
|
| 342 |
+
"grad_norm": 0.49588409066200256,
|
| 343 |
+
"learning_rate": 0.00010403120936280886,
|
| 344 |
+
"loss": 0.4706,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.01607084289931125,
|
| 349 |
+
"grad_norm": 0.6303009986877441,
|
| 350 |
+
"learning_rate": 0.00010203060918275482,
|
| 351 |
+
"loss": 0.6035,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.016398819285011478,
|
| 356 |
+
"grad_norm": 0.5468104481697083,
|
| 357 |
+
"learning_rate": 0.00010003000900270081,
|
| 358 |
+
"loss": 0.5344,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.016726795670711708,
|
| 363 |
+
"grad_norm": 0.5305197834968567,
|
| 364 |
+
"learning_rate": 9.802940882264679e-05,
|
| 365 |
+
"loss": 0.4671,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.017054772056411937,
|
| 370 |
+
"grad_norm": 0.6518839597702026,
|
| 371 |
+
"learning_rate": 9.602880864259278e-05,
|
| 372 |
+
"loss": 0.5079,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.017382748442112166,
|
| 377 |
+
"grad_norm": 0.6107843518257141,
|
| 378 |
+
"learning_rate": 9.402820846253876e-05,
|
| 379 |
+
"loss": 0.5327,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.0177107248278124,
|
| 384 |
+
"grad_norm": 0.550966739654541,
|
| 385 |
+
"learning_rate": 9.202760828248476e-05,
|
| 386 |
+
"loss": 0.5254,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.01803870121351263,
|
| 391 |
+
"grad_norm": 0.5233923196792603,
|
| 392 |
+
"learning_rate": 9.002700810243074e-05,
|
| 393 |
+
"loss": 0.5566,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.018366677599212858,
|
| 398 |
+
"grad_norm": 0.44530218839645386,
|
| 399 |
+
"learning_rate": 8.802640792237672e-05,
|
| 400 |
+
"loss": 0.5025,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.018694653984913087,
|
| 405 |
+
"grad_norm": 0.6728590130805969,
|
| 406 |
+
"learning_rate": 8.602580774232271e-05,
|
| 407 |
+
"loss": 0.5202,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.019022630370613317,
|
| 412 |
+
"grad_norm": 0.5043520927429199,
|
| 413 |
+
"learning_rate": 8.402520756226868e-05,
|
| 414 |
+
"loss": 0.5248,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.019350606756313546,
|
| 419 |
+
"grad_norm": 0.48946598172187805,
|
| 420 |
+
"learning_rate": 8.202460738221467e-05,
|
| 421 |
+
"loss": 0.5131,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.019678583142013775,
|
| 426 |
+
"grad_norm": 0.5078890323638916,
|
| 427 |
+
"learning_rate": 8.002400720216065e-05,
|
| 428 |
+
"loss": 0.5142,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.020006559527714005,
|
| 433 |
+
"grad_norm": 0.48533907532691956,
|
| 434 |
+
"learning_rate": 7.802340702210663e-05,
|
| 435 |
+
"loss": 0.5045,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.020334535913414234,
|
| 440 |
+
"grad_norm": 0.5382115840911865,
|
| 441 |
+
"learning_rate": 7.602280684205261e-05,
|
| 442 |
+
"loss": 0.5858,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.020662512299114463,
|
| 447 |
+
"grad_norm": 0.5158064961433411,
|
| 448 |
+
"learning_rate": 7.402220666199861e-05,
|
| 449 |
+
"loss": 0.525,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.020990488684814693,
|
| 454 |
+
"grad_norm": 0.5392168164253235,
|
| 455 |
+
"learning_rate": 7.202160648194459e-05,
|
| 456 |
+
"loss": 0.5439,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.021318465070514922,
|
| 461 |
+
"grad_norm": 0.5060271620750427,
|
| 462 |
+
"learning_rate": 7.002100630189057e-05,
|
| 463 |
+
"loss": 0.5141,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.02164644145621515,
|
| 468 |
+
"grad_norm": 0.5331196188926697,
|
| 469 |
+
"learning_rate": 6.802040612183656e-05,
|
| 470 |
+
"loss": 0.522,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.02197441784191538,
|
| 475 |
+
"grad_norm": 0.49599310755729675,
|
| 476 |
+
"learning_rate": 6.601980594178254e-05,
|
| 477 |
+
"loss": 0.5154,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.02230239422761561,
|
| 482 |
+
"grad_norm": 0.5665214657783508,
|
| 483 |
+
"learning_rate": 6.401920576172852e-05,
|
| 484 |
+
"loss": 0.5761,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.022630370613315843,
|
| 489 |
+
"grad_norm": 0.5102436542510986,
|
| 490 |
+
"learning_rate": 6.20186055816745e-05,
|
| 491 |
+
"loss": 0.5483,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.022958346999016072,
|
| 496 |
+
"grad_norm": 0.512220025062561,
|
| 497 |
+
"learning_rate": 6.001800540162049e-05,
|
| 498 |
+
"loss": 0.5357,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.0232863233847163,
|
| 503 |
+
"grad_norm": 0.5633516311645508,
|
| 504 |
+
"learning_rate": 5.801740522156648e-05,
|
| 505 |
+
"loss": 0.5605,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.02361429977041653,
|
| 510 |
+
"grad_norm": 0.4963136315345764,
|
| 511 |
+
"learning_rate": 5.601680504151246e-05,
|
| 512 |
+
"loss": 0.5617,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.02394227615611676,
|
| 517 |
+
"grad_norm": 0.6808409690856934,
|
| 518 |
+
"learning_rate": 5.4016204861458444e-05,
|
| 519 |
+
"loss": 0.5357,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.02427025254181699,
|
| 524 |
+
"grad_norm": 0.5417987704277039,
|
| 525 |
+
"learning_rate": 5.201560468140443e-05,
|
| 526 |
+
"loss": 0.5375,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.02459822892751722,
|
| 531 |
+
"grad_norm": 0.5455000996589661,
|
| 532 |
+
"learning_rate": 5.0015004501350405e-05,
|
| 533 |
+
"loss": 0.5362,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.02492620531321745,
|
| 538 |
+
"grad_norm": 0.49334102869033813,
|
| 539 |
+
"learning_rate": 4.801440432129639e-05,
|
| 540 |
+
"loss": 0.5081,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.025254181698917678,
|
| 545 |
+
"grad_norm": 0.4684203863143921,
|
| 546 |
+
"learning_rate": 4.601380414124238e-05,
|
| 547 |
+
"loss": 0.5061,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.025582158084617907,
|
| 552 |
+
"grad_norm": 0.5156847834587097,
|
| 553 |
+
"learning_rate": 4.401320396118836e-05,
|
| 554 |
+
"loss": 0.5231,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.025910134470318136,
|
| 559 |
+
"grad_norm": 0.5413920283317566,
|
| 560 |
+
"learning_rate": 4.201260378113434e-05,
|
| 561 |
+
"loss": 0.5849,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.026238110856018366,
|
| 566 |
+
"grad_norm": 0.5093967914581299,
|
| 567 |
+
"learning_rate": 4.0012003601080326e-05,
|
| 568 |
+
"loss": 0.497,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.026566087241718595,
|
| 573 |
+
"grad_norm": 0.46262067556381226,
|
| 574 |
+
"learning_rate": 3.801140342102631e-05,
|
| 575 |
+
"loss": 0.5364,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.026894063627418825,
|
| 580 |
+
"grad_norm": 0.6614812612533569,
|
| 581 |
+
"learning_rate": 3.6010803240972294e-05,
|
| 582 |
+
"loss": 0.4989,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.027222040013119054,
|
| 587 |
+
"grad_norm": 0.4883507788181305,
|
| 588 |
+
"learning_rate": 3.401020306091828e-05,
|
| 589 |
+
"loss": 0.5443,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.027550016398819287,
|
| 594 |
+
"grad_norm": 0.5155258774757385,
|
| 595 |
+
"learning_rate": 3.200960288086426e-05,
|
| 596 |
+
"loss": 0.5346,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.027877992784519516,
|
| 601 |
+
"grad_norm": 0.46190178394317627,
|
| 602 |
+
"learning_rate": 3.0009002700810245e-05,
|
| 603 |
+
"loss": 0.5019,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.028205969170219745,
|
| 608 |
+
"grad_norm": 0.6695094108581543,
|
| 609 |
+
"learning_rate": 2.800840252075623e-05,
|
| 610 |
+
"loss": 0.523,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.028533945555919975,
|
| 615 |
+
"grad_norm": 0.4700133204460144,
|
| 616 |
+
"learning_rate": 2.6007802340702216e-05,
|
| 617 |
+
"loss": 0.5288,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.028861921941620204,
|
| 622 |
+
"grad_norm": 0.5050464272499084,
|
| 623 |
+
"learning_rate": 2.4007202160648196e-05,
|
| 624 |
+
"loss": 0.4874,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.029189898327320433,
|
| 629 |
+
"grad_norm": 0.5086123943328857,
|
| 630 |
+
"learning_rate": 2.200660198059418e-05,
|
| 631 |
+
"loss": 0.557,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.029517874713020663,
|
| 636 |
+
"grad_norm": 0.48082900047302246,
|
| 637 |
+
"learning_rate": 2.0006001800540163e-05,
|
| 638 |
+
"loss": 0.5176,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.029845851098720892,
|
| 643 |
+
"grad_norm": 0.4965345561504364,
|
| 644 |
+
"learning_rate": 1.8005401620486147e-05,
|
| 645 |
+
"loss": 0.5113,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.03017382748442112,
|
| 650 |
+
"grad_norm": 0.49173784255981445,
|
| 651 |
+
"learning_rate": 1.600480144043213e-05,
|
| 652 |
+
"loss": 0.491,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.03050180387012135,
|
| 657 |
+
"grad_norm": 0.502700924873352,
|
| 658 |
+
"learning_rate": 1.4004201260378114e-05,
|
| 659 |
+
"loss": 0.522,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.03082978025582158,
|
| 664 |
+
"grad_norm": 0.5490285158157349,
|
| 665 |
+
"learning_rate": 1.2003601080324098e-05,
|
| 666 |
+
"loss": 0.4995,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.03115775664152181,
|
| 671 |
+
"grad_norm": 0.48369327187538147,
|
| 672 |
+
"learning_rate": 1.0003000900270082e-05,
|
| 673 |
+
"loss": 0.5166,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.03148573302722204,
|
| 678 |
+
"grad_norm": 0.6655266880989075,
|
| 679 |
+
"learning_rate": 8.002400720216065e-06,
|
| 680 |
+
"loss": 0.5704,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.03181370941292227,
|
| 685 |
+
"grad_norm": 0.4279884994029999,
|
| 686 |
+
"learning_rate": 6.001800540162049e-06,
|
| 687 |
+
"loss": 0.4882,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.0321416857986225,
|
| 692 |
+
"grad_norm": 0.5384986400604248,
|
| 693 |
+
"learning_rate": 4.001200360108033e-06,
|
| 694 |
+
"loss": 0.5809,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.03246966218432273,
|
| 699 |
+
"grad_norm": 0.4719259738922119,
|
| 700 |
+
"learning_rate": 2.0006001800540163e-06,
|
| 701 |
+
"loss": 0.5762,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.032797638570022956,
|
| 706 |
+
"grad_norm": 0.5332183241844177,
|
| 707 |
+
"learning_rate": 0.0,
|
| 708 |
+
"loss": 0.4694,
|
| 709 |
+
"step": 100
|
| 710 |
+
}
|
| 711 |
+
],
|
| 712 |
+
"logging_steps": 1,
|
| 713 |
+
"max_steps": 100,
|
| 714 |
+
"num_input_tokens_seen": 0,
|
| 715 |
+
"num_train_epochs": 1,
|
| 716 |
+
"save_steps": 500,
|
| 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": true
|
| 725 |
+
},
|
| 726 |
+
"attributes": {}
|
| 727 |
+
}
|
| 728 |
+
},
|
| 729 |
+
"total_flos": 6.686572160719258e+16,
|
| 730 |
+
"train_batch_size": 8,
|
| 731 |
+
"trial_name": null,
|
| 732 |
+
"trial_params": null
|
| 733 |
+
}
|