LLAMA2-446m / trainer_state.json
ccore's picture
Upload folder using huggingface_hub
1a2a470
raw
history blame
9.63 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9999916130601427,
"eval_steps": 500,
"global_step": 3726,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 9.86580783682233e-05,
"loss": 3.6768,
"step": 50
},
{
"epoch": 0.03,
"learning_rate": 9.731615673644659e-05,
"loss": 3.2865,
"step": 100
},
{
"epoch": 0.04,
"learning_rate": 9.597423510466989e-05,
"loss": 3.1518,
"step": 150
},
{
"epoch": 0.05,
"learning_rate": 9.463231347289319e-05,
"loss": 3.1091,
"step": 200
},
{
"epoch": 0.07,
"learning_rate": 9.329039184111649e-05,
"loss": 3.0456,
"step": 250
},
{
"epoch": 0.08,
"learning_rate": 9.194847020933978e-05,
"loss": 3.0357,
"step": 300
},
{
"epoch": 0.09,
"learning_rate": 9.060654857756307e-05,
"loss": 2.9846,
"step": 350
},
{
"epoch": 0.11,
"learning_rate": 8.926462694578636e-05,
"loss": 2.9834,
"step": 400
},
{
"epoch": 0.12,
"learning_rate": 8.792270531400967e-05,
"loss": 2.9385,
"step": 450
},
{
"epoch": 0.13,
"learning_rate": 8.658078368223296e-05,
"loss": 2.8967,
"step": 500
},
{
"epoch": 0.15,
"learning_rate": 8.523886205045626e-05,
"loss": 2.886,
"step": 550
},
{
"epoch": 0.16,
"learning_rate": 8.389694041867955e-05,
"loss": 2.8671,
"step": 600
},
{
"epoch": 0.17,
"learning_rate": 8.255501878690284e-05,
"loss": 2.8547,
"step": 650
},
{
"epoch": 0.19,
"learning_rate": 8.121309715512614e-05,
"loss": 2.8396,
"step": 700
},
{
"epoch": 0.2,
"learning_rate": 7.987117552334944e-05,
"loss": 2.8531,
"step": 750
},
{
"epoch": 0.21,
"learning_rate": 7.852925389157274e-05,
"loss": 2.8196,
"step": 800
},
{
"epoch": 0.23,
"learning_rate": 7.718733225979603e-05,
"loss": 2.7921,
"step": 850
},
{
"epoch": 0.24,
"learning_rate": 7.584541062801933e-05,
"loss": 2.7596,
"step": 900
},
{
"epoch": 0.25,
"learning_rate": 7.450348899624262e-05,
"loss": 2.7918,
"step": 950
},
{
"epoch": 0.27,
"learning_rate": 7.316156736446593e-05,
"loss": 2.7553,
"step": 1000
},
{
"epoch": 0.28,
"learning_rate": 7.181964573268921e-05,
"loss": 2.7914,
"step": 1050
},
{
"epoch": 0.3,
"learning_rate": 7.047772410091251e-05,
"loss": 2.7924,
"step": 1100
},
{
"epoch": 0.31,
"learning_rate": 6.91358024691358e-05,
"loss": 2.7823,
"step": 1150
},
{
"epoch": 0.32,
"learning_rate": 6.77938808373591e-05,
"loss": 2.7437,
"step": 1200
},
{
"epoch": 0.34,
"learning_rate": 6.64519592055824e-05,
"loss": 2.7404,
"step": 1250
},
{
"epoch": 0.35,
"learning_rate": 6.51100375738057e-05,
"loss": 2.7318,
"step": 1300
},
{
"epoch": 0.36,
"learning_rate": 6.376811594202898e-05,
"loss": 2.708,
"step": 1350
},
{
"epoch": 0.38,
"learning_rate": 6.242619431025228e-05,
"loss": 2.7579,
"step": 1400
},
{
"epoch": 0.39,
"learning_rate": 6.108427267847558e-05,
"loss": 2.7037,
"step": 1450
},
{
"epoch": 0.4,
"learning_rate": 5.9742351046698876e-05,
"loss": 2.7326,
"step": 1500
},
{
"epoch": 0.42,
"learning_rate": 5.8400429414922176e-05,
"loss": 2.7252,
"step": 1550
},
{
"epoch": 0.43,
"learning_rate": 5.705850778314546e-05,
"loss": 2.7263,
"step": 1600
},
{
"epoch": 0.44,
"learning_rate": 5.571658615136877e-05,
"loss": 2.6944,
"step": 1650
},
{
"epoch": 0.46,
"learning_rate": 5.4374664519592054e-05,
"loss": 2.7292,
"step": 1700
},
{
"epoch": 0.47,
"learning_rate": 5.3032742887815354e-05,
"loss": 2.7156,
"step": 1750
},
{
"epoch": 0.48,
"learning_rate": 5.1690821256038647e-05,
"loss": 2.6852,
"step": 1800
},
{
"epoch": 0.5,
"learning_rate": 5.0348899624261946e-05,
"loss": 2.6922,
"step": 1850
},
{
"epoch": 0.51,
"learning_rate": 4.9006977992485246e-05,
"loss": 2.7217,
"step": 1900
},
{
"epoch": 0.52,
"learning_rate": 4.766505636070854e-05,
"loss": 2.6923,
"step": 1950
},
{
"epoch": 0.54,
"learning_rate": 4.632313472893183e-05,
"loss": 2.6993,
"step": 2000
},
{
"epoch": 0.55,
"learning_rate": 4.498121309715513e-05,
"loss": 2.7156,
"step": 2050
},
{
"epoch": 0.56,
"learning_rate": 4.3639291465378424e-05,
"loss": 2.6932,
"step": 2100
},
{
"epoch": 0.58,
"learning_rate": 4.2297369833601716e-05,
"loss": 2.714,
"step": 2150
},
{
"epoch": 0.59,
"learning_rate": 4.0955448201825016e-05,
"loss": 2.6921,
"step": 2200
},
{
"epoch": 0.6,
"learning_rate": 3.961352657004831e-05,
"loss": 2.6949,
"step": 2250
},
{
"epoch": 0.62,
"learning_rate": 3.82716049382716e-05,
"loss": 2.694,
"step": 2300
},
{
"epoch": 0.63,
"learning_rate": 3.69296833064949e-05,
"loss": 2.6781,
"step": 2350
},
{
"epoch": 0.64,
"learning_rate": 3.55877616747182e-05,
"loss": 2.6932,
"step": 2400
},
{
"epoch": 0.66,
"learning_rate": 3.4245840042941493e-05,
"loss": 2.6748,
"step": 2450
},
{
"epoch": 0.67,
"learning_rate": 3.290391841116479e-05,
"loss": 2.6861,
"step": 2500
},
{
"epoch": 0.68,
"learning_rate": 3.1561996779388086e-05,
"loss": 2.6891,
"step": 2550
},
{
"epoch": 0.7,
"learning_rate": 3.0220075147611382e-05,
"loss": 2.7091,
"step": 2600
},
{
"epoch": 0.71,
"learning_rate": 2.8878153515834678e-05,
"loss": 2.6993,
"step": 2650
},
{
"epoch": 0.72,
"learning_rate": 2.753623188405797e-05,
"loss": 2.6827,
"step": 2700
},
{
"epoch": 0.74,
"learning_rate": 2.6194310252281267e-05,
"loss": 2.6876,
"step": 2750
},
{
"epoch": 0.75,
"learning_rate": 2.4852388620504563e-05,
"loss": 2.6812,
"step": 2800
},
{
"epoch": 0.76,
"learning_rate": 2.351046698872786e-05,
"loss": 2.6936,
"step": 2850
},
{
"epoch": 0.78,
"learning_rate": 2.2168545356951156e-05,
"loss": 2.6579,
"step": 2900
},
{
"epoch": 0.79,
"learning_rate": 2.0826623725174452e-05,
"loss": 2.6786,
"step": 2950
},
{
"epoch": 0.81,
"learning_rate": 1.9484702093397748e-05,
"loss": 2.6974,
"step": 3000
},
{
"epoch": 0.82,
"learning_rate": 1.814278046162104e-05,
"loss": 2.6818,
"step": 3050
},
{
"epoch": 0.83,
"learning_rate": 1.6800858829844337e-05,
"loss": 2.6742,
"step": 3100
},
{
"epoch": 0.85,
"learning_rate": 1.5458937198067633e-05,
"loss": 2.6748,
"step": 3150
},
{
"epoch": 0.86,
"learning_rate": 1.4117015566290927e-05,
"loss": 2.6771,
"step": 3200
},
{
"epoch": 0.87,
"learning_rate": 1.2775093934514227e-05,
"loss": 2.6895,
"step": 3250
},
{
"epoch": 0.89,
"learning_rate": 1.143317230273752e-05,
"loss": 2.6663,
"step": 3300
},
{
"epoch": 0.9,
"learning_rate": 1.0091250670960816e-05,
"loss": 2.6485,
"step": 3350
},
{
"epoch": 0.91,
"learning_rate": 8.749329039184112e-06,
"loss": 2.6825,
"step": 3400
},
{
"epoch": 0.93,
"learning_rate": 7.4074074074074075e-06,
"loss": 2.6832,
"step": 3450
},
{
"epoch": 0.94,
"learning_rate": 6.065485775630704e-06,
"loss": 2.7134,
"step": 3500
},
{
"epoch": 0.95,
"learning_rate": 4.723564143853999e-06,
"loss": 2.6631,
"step": 3550
},
{
"epoch": 0.97,
"learning_rate": 3.3816425120772947e-06,
"loss": 2.6862,
"step": 3600
},
{
"epoch": 0.98,
"learning_rate": 2.0397208803005905e-06,
"loss": 2.6663,
"step": 3650
},
{
"epoch": 0.99,
"learning_rate": 6.977992485238862e-07,
"loss": 2.6592,
"step": 3700
},
{
"epoch": 1.0,
"step": 3726,
"total_flos": 2.4427952859827405e+17,
"train_loss": 2.7717256223521947,
"train_runtime": 3102.8464,
"train_samples_per_second": 38.427,
"train_steps_per_second": 1.201
}
],
"logging_steps": 50,
"max_steps": 3726,
"num_train_epochs": 1,
"save_steps": 5000,
"total_flos": 2.4427952859827405e+17,
"trial_name": null,
"trial_params": null
}