eminem / trainer_state.json
AlekseyKorshuk's picture
huggingartists
ad9a638
raw
history blame
11.8 kB
{
"best_metric": 0.21638885140419006,
"best_model_checkpoint": "output/eminem/checkpoint-464",
"epoch": 1.0,
"global_step": 464,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.956064306819415e-05,
"loss": 0.507,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 4.7311949670299195e-05,
"loss": 0.5191,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 4.5088294415386574e-05,
"loss": 0.558,
"step": 15
},
{
"epoch": 0.04,
"learning_rate": 4.289229267674363e-05,
"loss": 0.5324,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 4.0726527302713256e-05,
"loss": 0.4462,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 3.859354557885036e-05,
"loss": 0.5269,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 3.649585623190601e-05,
"loss": 0.5423,
"step": 35
},
{
"epoch": 0.09,
"learning_rate": 3.4435926479171496e-05,
"loss": 0.5446,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 3.241617912662684e-05,
"loss": 0.4715,
"step": 45
},
{
"epoch": 0.11,
"learning_rate": 3.043898971933976e-05,
"loss": 0.531,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 2.8506683747450634e-05,
"loss": 0.553,
"step": 55
},
{
"epoch": 0.13,
"learning_rate": 2.662153391101519e-05,
"loss": 0.51,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 2.478575744695336e-05,
"loss": 0.5793,
"step": 65
},
{
"epoch": 0.15,
"learning_rate": 2.3001513521221825e-05,
"loss": 0.4464,
"step": 70
},
{
"epoch": 0.16,
"learning_rate": 2.1270900689296354e-05,
"loss": 0.5047,
"step": 75
},
{
"epoch": 0.17,
"learning_rate": 1.9595954427926873e-05,
"loss": 0.4798,
"step": 80
},
{
"epoch": 0.19,
"learning_rate": 1.7978644741096808e-05,
"loss": 0.5132,
"step": 85
},
{
"epoch": 0.2,
"learning_rate": 1.6420873842987938e-05,
"loss": 0.5137,
"step": 90
},
{
"epoch": 0.21,
"learning_rate": 1.492447392066504e-05,
"loss": 0.5699,
"step": 95
},
{
"epoch": 0.22,
"learning_rate": 1.3491204979137085e-05,
"loss": 0.5101,
"step": 100
},
{
"epoch": 0.23,
"learning_rate": 1.2122752771309161e-05,
"loss": 0.4993,
"step": 105
},
{
"epoch": 0.24,
"learning_rate": 1.0820726815267185e-05,
"loss": 0.5695,
"step": 110
},
{
"epoch": 0.25,
"learning_rate": 9.586658501232076e-06,
"loss": 0.5484,
"step": 115
},
{
"epoch": 0.26,
"learning_rate": 8.421999290393203e-06,
"loss": 0.5208,
"step": 120
},
{
"epoch": 0.27,
"learning_rate": 7.32811900776668e-06,
"loss": 0.443,
"step": 125
},
{
"epoch": 0.28,
"learning_rate": 6.306304231055696e-06,
"loss": 0.4872,
"step": 130
},
{
"epoch": 0.29,
"learning_rate": 5.357756777433853e-06,
"loss": 0.5517,
"step": 135
},
{
"epoch": 0.31,
"learning_rate": 4.48359229001547e-06,
"loss": 0.5653,
"step": 140
},
{
"epoch": 0.32,
"learning_rate": 3.6848389256806803e-06,
"loss": 0.5335,
"step": 145
},
{
"epoch": 0.33,
"learning_rate": 2.9624361458015066e-06,
"loss": 0.5232,
"step": 150
},
{
"epoch": 0.34,
"learning_rate": 2.3172336112814763e-06,
"loss": 0.4785,
"step": 155
},
{
"epoch": 0.35,
"learning_rate": 1.7499901832211055e-06,
"loss": 0.5124,
"step": 160
},
{
"epoch": 0.36,
"learning_rate": 1.2613730303773645e-06,
"loss": 0.4994,
"step": 165
},
{
"epoch": 0.37,
"learning_rate": 8.519568444636254e-07,
"loss": 0.4736,
"step": 170
},
{
"epoch": 0.38,
"learning_rate": 5.222231642240317e-07,
"loss": 0.5123,
"step": 175
},
{
"epoch": 0.39,
"learning_rate": 2.7255980906336875e-07,
"loss": 0.5742,
"step": 180
},
{
"epoch": 0.4,
"learning_rate": 1.0326042291343123e-07,
"loss": 0.5612,
"step": 185
},
{
"epoch": 0.41,
"learning_rate": 1.4524128858102236e-08,
"loss": 0.4673,
"step": 190
},
{
"epoch": 0.43,
"learning_rate": 6.455294934120847e-09,
"loss": 0.5373,
"step": 195
},
{
"epoch": 0.44,
"learning_rate": 7.906341137730022e-08,
"loss": 0.5127,
"step": 200
},
{
"epoch": 0.45,
"learning_rate": 2.3226307946064745e-07,
"loss": 0.5527,
"step": 205
},
{
"epoch": 0.46,
"learning_rate": 4.658741119363803e-07,
"loss": 0.518,
"step": 210
},
{
"epoch": 0.47,
"learning_rate": 7.796217449653221e-07,
"loss": 0.5127,
"step": 215
},
{
"epoch": 0.48,
"learning_rate": 1.1731369612840247e-06,
"loss": 0.4868,
"step": 220
},
{
"epoch": 0.49,
"learning_rate": 1.64595692422509e-06,
"loss": 0.4982,
"step": 225
},
{
"epoch": 0.5,
"learning_rate": 2.197525522090362e-06,
"loss": 0.5943,
"step": 230
},
{
"epoch": 0.51,
"learning_rate": 2.827194022223598e-06,
"loss": 0.4951,
"step": 235
},
{
"epoch": 0.52,
"learning_rate": 3.5342218340266394e-06,
"loss": 0.5133,
"step": 240
},
{
"epoch": 0.53,
"learning_rate": 4.317777380010967e-06,
"loss": 0.5839,
"step": 245
},
{
"epoch": 0.55,
"learning_rate": 5.176939073863253e-06,
"loss": 0.5613,
"step": 250
},
{
"epoch": 0.56,
"learning_rate": 6.1106964043817805e-06,
"loss": 0.5325,
"step": 255
},
{
"epoch": 0.57,
"learning_rate": 7.11795112399628e-06,
"loss": 0.4642,
"step": 260
},
{
"epoch": 0.58,
"learning_rate": 8.197518540483775e-06,
"loss": 0.5172,
"step": 265
},
{
"epoch": 0.59,
"learning_rate": 9.348128910351998e-06,
"loss": 0.5789,
"step": 270
},
{
"epoch": 0.6,
"learning_rate": 1.056842893226566e-05,
"loss": 0.477,
"step": 275
},
{
"epoch": 0.61,
"learning_rate": 1.1856983338742235e-05,
"loss": 0.4877,
"step": 280
},
{
"epoch": 0.62,
"learning_rate": 1.3212276584252036e-05,
"loss": 0.4706,
"step": 285
},
{
"epoch": 0.63,
"learning_rate": 1.4632714627747902e-05,
"loss": 0.5367,
"step": 290
},
{
"epoch": 0.64,
"learning_rate": 1.6116626807507566e-05,
"loss": 0.5493,
"step": 295
},
{
"epoch": 0.66,
"learning_rate": 1.766226780609932e-05,
"loss": 0.444,
"step": 300
},
{
"epoch": 0.67,
"learning_rate": 1.926781970315512e-05,
"loss": 0.4818,
"step": 305
},
{
"epoch": 0.68,
"learning_rate": 2.093139411352953e-05,
"loss": 0.4411,
"step": 310
},
{
"epoch": 0.69,
"learning_rate": 2.265103440835135e-05,
"loss": 0.5486,
"step": 315
},
{
"epoch": 0.7,
"learning_rate": 2.442471801632022e-05,
"loss": 0.4986,
"step": 320
},
{
"epoch": 0.71,
"learning_rate": 2.625035880258298e-05,
"loss": 0.504,
"step": 325
},
{
"epoch": 0.72,
"learning_rate": 2.8125809522355368e-05,
"loss": 0.507,
"step": 330
},
{
"epoch": 0.73,
"learning_rate": 3.0048864346425834e-05,
"loss": 0.5913,
"step": 335
},
{
"epoch": 0.74,
"learning_rate": 3.201726145556417e-05,
"loss": 0.5699,
"step": 340
},
{
"epoch": 0.75,
"learning_rate": 3.4028685700775186e-05,
"loss": 0.5309,
"step": 345
},
{
"epoch": 0.76,
"learning_rate": 3.608077132629301e-05,
"loss": 0.5089,
"step": 350
},
{
"epoch": 0.78,
"learning_rate": 3.817110475208252e-05,
"loss": 0.5464,
"step": 355
},
{
"epoch": 0.79,
"learning_rate": 4.029722741258908e-05,
"loss": 0.4744,
"step": 360
},
{
"epoch": 0.8,
"learning_rate": 4.2456638648414265e-05,
"loss": 0.5028,
"step": 365
},
{
"epoch": 0.81,
"learning_rate": 4.46467986474833e-05,
"loss": 0.5464,
"step": 370
},
{
"epoch": 0.82,
"learning_rate": 4.6865131432272725e-05,
"loss": 0.4575,
"step": 375
},
{
"epoch": 0.83,
"learning_rate": 4.910902788956352e-05,
"loss": 0.531,
"step": 380
},
{
"epoch": 0.84,
"learning_rate": 5.1375848839185514e-05,
"loss": 0.5465,
"step": 385
},
{
"epoch": 0.85,
"learning_rate": 5.366292813810878e-05,
"loss": 0.4969,
"step": 390
},
{
"epoch": 0.86,
"learning_rate": 5.596757581624737e-05,
"loss": 0.5017,
"step": 395
},
{
"epoch": 0.87,
"learning_rate": 5.8287081240304644e-05,
"loss": 0.5094,
"step": 400
},
{
"epoch": 0.88,
"learning_rate": 6.061871630190278e-05,
"loss": 0.537,
"step": 405
},
{
"epoch": 0.9,
"learning_rate": 6.295973862627538e-05,
"loss": 0.4951,
"step": 410
},
{
"epoch": 0.91,
"learning_rate": 6.53073947977395e-05,
"loss": 0.5339,
"step": 415
},
{
"epoch": 0.92,
"learning_rate": 6.765892359814399e-05,
"loss": 0.5404,
"step": 420
},
{
"epoch": 0.93,
"learning_rate": 7.001155925451613e-05,
"loss": 0.4823,
"step": 425
},
{
"epoch": 0.94,
"learning_rate": 7.236253469203617e-05,
"loss": 0.5194,
"step": 430
},
{
"epoch": 0.95,
"learning_rate": 7.470908478857246e-05,
"loss": 0.5532,
"step": 435
},
{
"epoch": 0.96,
"learning_rate": 7.704844962689829e-05,
"loss": 0.5701,
"step": 440
},
{
"epoch": 0.97,
"learning_rate": 7.937787774079682e-05,
"loss": 0.5016,
"step": 445
},
{
"epoch": 0.98,
"learning_rate": 8.169462935122601e-05,
"loss": 0.6026,
"step": 450
},
{
"epoch": 0.99,
"learning_rate": 8.399597958872837e-05,
"loss": 0.5809,
"step": 455
},
{
"epoch": 1.0,
"eval_loss": 0.2587912380695343,
"eval_runtime": 28.9508,
"eval_samples_per_second": 21.243,
"eval_steps_per_second": 2.66,
"step": 458
},
{
"epoch": 0.99,
"learning_rate": 8.656213912947592e-06,
"loss": 0.6179,
"step": 460
},
{
"epoch": 1.0,
"eval_loss": 0.21638885140419006,
"eval_runtime": 7.4743,
"eval_samples_per_second": 75.994,
"eval_steps_per_second": 9.499,
"step": 464
}
],
"max_steps": 7424,
"num_train_epochs": 16,
"total_flos": 484304781312000.0,
"trial_name": null,
"trial_params": null
}