aikko / trainer_state.json
AlekseyKorshuk's picture
huggingartists
7824977
{
"best_metric": 1.687872052192688,
"best_model_checkpoint": "output/aikko/checkpoint-251",
"epoch": 1.0,
"global_step": 251,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 0.0001370657096871596,
"loss": 2.6148,
"step": 5
},
{
"epoch": 0.04,
"learning_rate": 0.00013666336451797142,
"loss": 2.2277,
"step": 10
},
{
"epoch": 0.06,
"learning_rate": 0.0001359945397419585,
"loss": 2.1213,
"step": 15
},
{
"epoch": 0.08,
"learning_rate": 0.00013506185392146502,
"loss": 2.0681,
"step": 20
},
{
"epoch": 0.1,
"learning_rate": 0.00013386895867954104,
"loss": 2.0275,
"step": 25
},
{
"epoch": 0.12,
"learning_rate": 0.0001324205244032193,
"loss": 1.9672,
"step": 30
},
{
"epoch": 0.14,
"learning_rate": 0.00013072222195815859,
"loss": 1.9337,
"step": 35
},
{
"epoch": 0.16,
"learning_rate": 0.00012878070048624358,
"loss": 1.9796,
"step": 40
},
{
"epoch": 0.18,
"learning_rate": 0.00012660356137306759,
"loss": 1.8898,
"step": 45
},
{
"epoch": 0.2,
"learning_rate": 0.00012419932848721968,
"loss": 1.914,
"step": 50
},
{
"epoch": 0.22,
"learning_rate": 0.00012157741480789465,
"loss": 1.795,
"step": 55
},
{
"epoch": 0.24,
"learning_rate": 0.00011874808557148426,
"loss": 1.8209,
"step": 60
},
{
"epoch": 0.26,
"learning_rate": 0.00011572241808143707,
"loss": 1.8927,
"step": 65
},
{
"epoch": 0.28,
"learning_rate": 0.00011251225833873822,
"loss": 1.7561,
"step": 70
},
{
"epoch": 0.3,
"learning_rate": 0.00010913017466280793,
"loss": 1.8688,
"step": 75
},
{
"epoch": 0.32,
"learning_rate": 0.00010558940848440114,
"loss": 1.8271,
"step": 80
},
{
"epoch": 0.34,
"learning_rate": 0.00010190382250316192,
"loss": 1.8149,
"step": 85
},
{
"epoch": 0.36,
"learning_rate": 9.808784641280521e-05,
"loss": 1.8287,
"step": 90
},
{
"epoch": 0.38,
"learning_rate": 9.415642040642034e-05,
"loss": 1.8284,
"step": 95
},
{
"epoch": 0.4,
"learning_rate": 9.01249366830833e-05,
"loss": 1.8122,
"step": 100
},
{
"epoch": 0.42,
"learning_rate": 8.600917918478882e-05,
"loss": 1.7182,
"step": 105
},
{
"epoch": 0.44,
"learning_rate": 8.182526179964243e-05,
"loss": 1.7897,
"step": 110
},
{
"epoch": 0.46,
"learning_rate": 7.758956527325783e-05,
"loss": 1.7654,
"step": 115
},
{
"epoch": 0.48,
"learning_rate": 7.331867307536173e-05,
"loss": 1.8424,
"step": 120
},
{
"epoch": 0.5,
"learning_rate": 6.902930647269938e-05,
"loss": 1.831,
"step": 125
},
{
"epoch": 0.52,
"learning_rate": 6.473825906244161e-05,
"loss": 1.7846,
"step": 130
},
{
"epoch": 0.54,
"learning_rate": 6.046233102240582e-05,
"loss": 1.7197,
"step": 135
},
{
"epoch": 0.56,
"learning_rate": 5.621826333551266e-05,
"loss": 1.6044,
"step": 140
},
{
"epoch": 0.58,
"learning_rate": 5.202267224600086e-05,
"loss": 1.6492,
"step": 145
},
{
"epoch": 0.6,
"learning_rate": 4.78919842040151e-05,
"loss": 1.761,
"step": 150
},
{
"epoch": 0.62,
"learning_rate": 4.3842371553269784e-05,
"loss": 1.8122,
"step": 155
},
{
"epoch": 0.64,
"learning_rate": 3.98896892135827e-05,
"loss": 1.7144,
"step": 160
},
{
"epoch": 0.66,
"learning_rate": 3.604941260617706e-05,
"loss": 1.6734,
"step": 165
},
{
"epoch": 0.68,
"learning_rate": 3.2336577064784687e-05,
"loss": 1.7209,
"step": 170
},
{
"epoch": 0.7,
"learning_rate": 2.876571896976647e-05,
"loss": 1.6936,
"step": 175
},
{
"epoch": 0.72,
"learning_rate": 2.535081883571979e-05,
"loss": 1.7183,
"step": 180
},
{
"epoch": 0.74,
"learning_rate": 2.2105246575394433e-05,
"loss": 1.6964,
"step": 185
},
{
"epoch": 0.76,
"learning_rate": 1.9041709154218593e-05,
"loss": 1.6846,
"step": 190
},
{
"epoch": 0.78,
"learning_rate": 1.6172200840375557e-05,
"loss": 1.7342,
"step": 195
},
{
"epoch": 0.8,
"learning_rate": 1.350795624521122e-05,
"loss": 1.6946,
"step": 200
},
{
"epoch": 0.82,
"learning_rate": 1.1059406337826712e-05,
"loss": 1.5966,
"step": 205
},
{
"epoch": 0.84,
"learning_rate": 8.836137606066586e-06,
"loss": 1.6202,
"step": 210
},
{
"epoch": 0.86,
"learning_rate": 6.846854523794257e-06,
"loss": 1.6442,
"step": 215
},
{
"epoch": 0.88,
"learning_rate": 5.099345471401779e-06,
"loss": 1.7818,
"step": 220
},
{
"epoch": 0.9,
"learning_rate": 3.6004522429806e-06,
"loss": 1.5978,
"step": 225
},
{
"epoch": 0.92,
"learning_rate": 2.3560432595382876e-06,
"loss": 1.7863,
"step": 230
},
{
"epoch": 0.94,
"learning_rate": 1.3709905931356957e-06,
"loss": 1.7466,
"step": 235
},
{
"epoch": 0.96,
"learning_rate": 6.491508918992284e-07,
"loss": 1.6058,
"step": 240
},
{
"epoch": 0.98,
"learning_rate": 1.9335028058981832e-07,
"loss": 1.6482,
"step": 245
},
{
"epoch": 1.0,
"learning_rate": 5.373295845529613e-09,
"loss": 1.7028,
"step": 250
},
{
"epoch": 1.0,
"eval_loss": 1.687872052192688,
"eval_runtime": 16.5648,
"eval_samples_per_second": 22.518,
"eval_steps_per_second": 2.837,
"step": 251
}
],
"max_steps": 251,
"num_train_epochs": 1,
"total_flos": 262206554112000.0,
"trial_name": null,
"trial_params": null
}