Files
east8b-eg-lora-chunk-llama/last-checkpoint/trainer_state.json

434 lines
10 KiB
JSON
Raw Normal View History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.9965277777777777,
"eval_steps": 500,
"global_step": 1150,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.034722222222222224,
"grad_norm": 2.4707443714141846,
"learning_rate": 1.6379310344827587e-06,
"loss": 2.5661,
"step": 20
},
{
"epoch": 0.06944444444444445,
"grad_norm": 2.0097849369049072,
"learning_rate": 3.362068965517242e-06,
"loss": 2.3797,
"step": 40
},
{
"epoch": 0.10416666666666667,
"grad_norm": 1.9553152322769165,
"learning_rate": 5.086206896551724e-06,
"loss": 2.241,
"step": 60
},
{
"epoch": 0.1388888888888889,
"grad_norm": 1.8024173974990845,
"learning_rate": 6.810344827586207e-06,
"loss": 2.0635,
"step": 80
},
{
"epoch": 0.1736111111111111,
"grad_norm": 2.215123176574707,
"learning_rate": 8.53448275862069e-06,
"loss": 1.9409,
"step": 100
},
{
"epoch": 0.20833333333333334,
"grad_norm": 2.2895090579986572,
"learning_rate": 9.999793100349294e-06,
"loss": 1.9394,
"step": 120
},
{
"epoch": 0.24305555555555555,
"grad_norm": 2.780405044555664,
"learning_rate": 9.987843743451796e-06,
"loss": 1.8307,
"step": 140
},
{
"epoch": 0.2777777777777778,
"grad_norm": 2.183285713195801,
"learning_rate": 9.957553516178782e-06,
"loss": 1.7575,
"step": 160
},
{
"epoch": 0.3125,
"grad_norm": 2.269092082977295,
"learning_rate": 9.909033799150947e-06,
"loss": 1.8066,
"step": 180
},
{
"epoch": 0.3472222222222222,
"grad_norm": 2.379218578338623,
"learning_rate": 9.842463004902127e-06,
"loss": 1.7141,
"step": 200
},
{
"epoch": 0.3819444444444444,
"grad_norm": 2.7214298248291016,
"learning_rate": 9.758085921836076e-06,
"loss": 1.71,
"step": 220
},
{
"epoch": 0.4166666666666667,
"grad_norm": 2.449744462966919,
"learning_rate": 9.656212814111567e-06,
"loss": 1.7129,
"step": 240
},
{
"epoch": 0.4513888888888889,
"grad_norm": 3.360551118850708,
"learning_rate": 9.53721828076571e-06,
"loss": 1.6855,
"step": 260
},
{
"epoch": 0.4861111111111111,
"grad_norm": 3.03425669670105,
"learning_rate": 9.401539878270545e-06,
"loss": 1.6934,
"step": 280
},
{
"epoch": 0.5208333333333334,
"grad_norm": 3.087336301803589,
"learning_rate": 9.249676511588e-06,
"loss": 1.6952,
"step": 300
},
{
"epoch": 0.5555555555555556,
"grad_norm": 2.6650753021240234,
"learning_rate": 9.082186599639429e-06,
"loss": 1.6734,
"step": 320
},
{
"epoch": 0.5902777777777778,
"grad_norm": 2.713989496231079,
"learning_rate": 8.899686021935554e-06,
"loss": 1.6516,
"step": 340
},
{
"epoch": 0.625,
"grad_norm": 2.6662213802337646,
"learning_rate": 8.702845853917242e-06,
"loss": 1.6282,
"step": 360
},
{
"epoch": 0.6597222222222222,
"grad_norm": 2.6345484256744385,
"learning_rate": 8.492389899334572e-06,
"loss": 1.6137,
"step": 380
},
{
"epoch": 0.6944444444444444,
"grad_norm": 2.9651432037353516,
"learning_rate": 8.269092028737885e-06,
"loss": 1.6086,
"step": 400
},
{
"epoch": 0.7291666666666666,
"grad_norm": 2.811237335205078,
"learning_rate": 8.033773333867498e-06,
"loss": 1.5801,
"step": 420
},
{
"epoch": 0.7638888888888888,
"grad_norm": 2.8691394329071045,
"learning_rate": 7.78729910840572e-06,
"loss": 1.5673,
"step": 440
},
{
"epoch": 0.7986111111111112,
"grad_norm": 2.8986737728118896,
"learning_rate": 7.530575666193283e-06,
"loss": 1.5777,
"step": 460
},
{
"epoch": 0.8333333333333334,
"grad_norm": 3.5703928470611572,
"learning_rate": 7.26454700860997e-06,
"loss": 1.5584,
"step": 480
},
{
"epoch": 0.8680555555555556,
"grad_norm": 3.996887445449829,
"learning_rate": 6.990191353373876e-06,
"loss": 1.5649,
"step": 500
},
{
"epoch": 0.9027777777777778,
"grad_norm": 3.5608773231506348,
"learning_rate": 6.708517537523264e-06,
"loss": 1.518,
"step": 520
},
{
"epoch": 0.9375,
"grad_norm": 3.395473003387451,
"learning_rate": 6.420561307807713e-06,
"loss": 1.5457,
"step": 540
},
{
"epoch": 0.9722222222222222,
"grad_norm": 2.847874164581299,
"learning_rate": 6.12738151212918e-06,
"loss": 1.5655,
"step": 560
},
{
"epoch": 1.0069444444444444,
"grad_norm": 3.7648465633392334,
"learning_rate": 5.830056206037482e-06,
"loss": 1.5093,
"step": 580
},
{
"epoch": 1.0416666666666667,
"grad_norm": 3.106715440750122,
"learning_rate": 5.529678688597081e-06,
"loss": 1.4521,
"step": 600
},
{
"epoch": 1.0763888888888888,
"grad_norm": 3.3356199264526367,
"learning_rate": 5.2273534822017105e-06,
"loss": 1.461,
"step": 620
},
{
"epoch": 1.1111111111111112,
"grad_norm": 3.7472050189971924,
"learning_rate": 4.924192271119554e-06,
"loss": 1.4325,
"step": 640
},
{
"epoch": 1.1458333333333333,
"grad_norm": 3.0159659385681152,
"learning_rate": 4.621309813703385e-06,
"loss": 1.44,
"step": 660
},
{
"epoch": 1.1805555555555556,
"grad_norm": 4.520766258239746,
"learning_rate": 4.319819843296952e-06,
"loss": 1.4462,
"step": 680
},
{
"epoch": 1.2152777777777777,
"grad_norm": 3.7417140007019043,
"learning_rate": 4.020830972910433e-06,
"loss": 1.4551,
"step": 700
},
{
"epoch": 1.25,
"grad_norm": 3.331568479537964,
"learning_rate": 3.7254426187239567e-06,
"loss": 1.4033,
"step": 720
},
{
"epoch": 1.2847222222222223,
"grad_norm": 3.6196417808532715,
"learning_rate": 3.4347409574088896e-06,
"loss": 1.4319,
"step": 740
},
{
"epoch": 1.3194444444444444,
"grad_norm": 4.271605491638184,
"learning_rate": 3.149794932132331e-06,
"loss": 1.4318,
"step": 760
},
{
"epoch": 1.3541666666666667,
"grad_norm": 4.33790922164917,
"learning_rate": 2.871652321931161e-06,
"loss": 1.4123,
"step": 780
},
{
"epoch": 1.3888888888888888,
"grad_norm": 2.9214086532592773,
"learning_rate": 2.601335888909005e-06,
"loss": 1.4245,
"step": 800
},
{
"epoch": 1.4236111111111112,
"grad_norm": 3.770592212677002,
"learning_rate": 2.339839617423318e-06,
"loss": 1.3763,
"step": 820
},
{
"epoch": 1.4583333333333333,
"grad_norm": 3.762789726257324,
"learning_rate": 2.0881250590915316e-06,
"loss": 1.4043,
"step": 840
},
{
"epoch": 1.4930555555555556,
"grad_norm": 3.8584647178649902,
"learning_rate": 1.8471177970560712e-06,
"loss": 1.4174,
"step": 860
},
{
"epoch": 1.5277777777777777,
"grad_norm": 3.3003017902374268,
"learning_rate": 1.6177040425095664e-06,
"loss": 1.3779,
"step": 880
},
{
"epoch": 1.5625,
"grad_norm": 3.60571551322937,
"learning_rate": 1.40072737599522e-06,
"loss": 1.4303,
"step": 900
},
{
"epoch": 1.5972222222222223,
"grad_norm": 3.2599637508392334,
"learning_rate": 1.196985645464921e-06,
"loss": 1.3981,
"step": 920
},
{
"epoch": 1.6319444444444444,
"grad_norm": 4.389312267303467,
"learning_rate": 1.0072280325013185e-06,
"loss": 1.3968,
"step": 940
},
{
"epoch": 1.6666666666666665,
"grad_norm": 4.148715972900391,
"learning_rate": 8.321522974916968e-07,
"loss": 1.422,
"step": 960
},
{
"epoch": 1.7013888888888888,
"grad_norm": 4.725743770599365,
"learning_rate": 6.724022138834341e-07,
"loss": 1.3997,
"step": 980
},
{
"epoch": 1.7361111111111112,
"grad_norm": 3.287371873855591,
"learning_rate": 5.285652009556075e-07,
"loss": 1.3733,
"step": 1000
},
{
"epoch": 1.7708333333333335,
"grad_norm": 4.199480056762695,
"learning_rate": 4.0117016381130636e-07,
"loss": 1.3904,
"step": 1020
},
{
"epoch": 1.8055555555555556,
"grad_norm": 3.8616933822631836,
"learning_rate": 2.906855485332305e-07,
"loss": 1.3662,
"step": 1040
},
{
"epoch": 1.8402777777777777,
"grad_norm": 3.60861873626709,
"learning_rate": 1.975176196540185e-07,
"loss": 1.3885,
"step": 1060
},
{
"epoch": 1.875,
"grad_norm": 4.208493709564209,
"learning_rate": 1.2200896627521718e-07,
"loss": 1.3956,
"step": 1080
},
{
"epoch": 1.9097222222222223,
"grad_norm": 3.6999189853668213,
"learning_rate": 6.443724232808146e-08,
"loss": 1.3784,
"step": 1100
},
{
"epoch": 1.9444444444444444,
"grad_norm": 5.089156627655029,
"learning_rate": 2.501414560839577e-08,
"loss": 1.3576,
"step": 1120
},
{
"epoch": 1.9791666666666665,
"grad_norm": 3.6565749645233154,
"learning_rate": 3.884639339534202e-09,
"loss": 1.37,
"step": 1140
}
],
"logging_steps": 20,
"max_steps": 1152,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 50,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2.2986760990910054e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}