Grogros's picture
Training in progress, step 2000, checkpoint
5dd466f verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.12843565373747753,
"eval_steps": 500,
"global_step": 2000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.003210891343436938,
"grad_norm": 8.177531242370605,
"learning_rate": 4.000000000000001e-06,
"loss": 2.2773,
"step": 50
},
{
"epoch": 0.006421782686873876,
"grad_norm": 0.6572615504264832,
"learning_rate": 8.000000000000001e-06,
"loss": 0.2291,
"step": 100
},
{
"epoch": 0.009632674030310814,
"grad_norm": 0.6884369254112244,
"learning_rate": 1.2e-05,
"loss": 0.1612,
"step": 150
},
{
"epoch": 0.012843565373747753,
"grad_norm": 0.32834213972091675,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.1469,
"step": 200
},
{
"epoch": 0.016054456717184692,
"grad_norm": 0.3822823166847229,
"learning_rate": 2e-05,
"loss": 0.1699,
"step": 250
},
{
"epoch": 0.019265348060621627,
"grad_norm": 0.3158350884914398,
"learning_rate": 1.9975640502598243e-05,
"loss": 0.1448,
"step": 300
},
{
"epoch": 0.022476239404058566,
"grad_norm": 984.7891845703125,
"learning_rate": 1.9902680687415704e-05,
"loss": 0.1638,
"step": 350
},
{
"epoch": 0.025687130747495505,
"grad_norm": 0.2283572256565094,
"learning_rate": 1.9781476007338058e-05,
"loss": 0.1346,
"step": 400
},
{
"epoch": 0.028898022090932445,
"grad_norm": 0.24141617119312286,
"learning_rate": 1.961261695938319e-05,
"loss": 0.1347,
"step": 450
},
{
"epoch": 0.032108913434369384,
"grad_norm": 0.2759339213371277,
"learning_rate": 1.9396926207859085e-05,
"loss": 0.1269,
"step": 500
},
{
"epoch": 0.035319804777806316,
"grad_norm": 0.23010869324207306,
"learning_rate": 1.913545457642601e-05,
"loss": 0.1286,
"step": 550
},
{
"epoch": 0.038530696121243255,
"grad_norm": 0.22603748738765717,
"learning_rate": 1.8829475928589272e-05,
"loss": 0.1288,
"step": 600
},
{
"epoch": 0.041741587464680194,
"grad_norm": 0.20579472184181213,
"learning_rate": 1.848048096156426e-05,
"loss": 0.1253,
"step": 650
},
{
"epoch": 0.04495247880811713,
"grad_norm": 0.19978702068328857,
"learning_rate": 1.8090169943749477e-05,
"loss": 0.1231,
"step": 700
},
{
"epoch": 0.04816337015155407,
"grad_norm": 0.24899648129940033,
"learning_rate": 1.766044443118978e-05,
"loss": 0.1225,
"step": 750
},
{
"epoch": 0.05137426149499101,
"grad_norm": 0.2297627180814743,
"learning_rate": 1.7193398003386514e-05,
"loss": 0.1214,
"step": 800
},
{
"epoch": 0.05458515283842795,
"grad_norm": 0.1948223114013672,
"learning_rate": 1.6691306063588583e-05,
"loss": 0.1233,
"step": 850
},
{
"epoch": 0.05779604418186489,
"grad_norm": 0.2076019048690796,
"learning_rate": 1.6156614753256583e-05,
"loss": 0.1173,
"step": 900
},
{
"epoch": 0.06100693552530182,
"grad_norm": 0.1943691223859787,
"learning_rate": 1.5591929034707468e-05,
"loss": 0.121,
"step": 950
},
{
"epoch": 0.06421782686873877,
"grad_norm": 0.1736869364976883,
"learning_rate": 1.5000000000000002e-05,
"loss": 0.1181,
"step": 1000
},
{
"epoch": 0.0674287182121757,
"grad_norm": 0.2161734253168106,
"learning_rate": 1.4383711467890776e-05,
"loss": 0.1145,
"step": 1050
},
{
"epoch": 0.07063960955561263,
"grad_norm": 0.21289649605751038,
"learning_rate": 1.3746065934159123e-05,
"loss": 0.1148,
"step": 1100
},
{
"epoch": 0.07385050089904957,
"grad_norm": 0.21228502690792084,
"learning_rate": 1.3090169943749475e-05,
"loss": 0.1176,
"step": 1150
},
{
"epoch": 0.07706139224248651,
"grad_norm": 0.16968905925750732,
"learning_rate": 1.2419218955996677e-05,
"loss": 0.1136,
"step": 1200
},
{
"epoch": 0.08027228358592345,
"grad_norm": 0.20333637297153473,
"learning_rate": 1.1736481776669307e-05,
"loss": 0.1135,
"step": 1250
},
{
"epoch": 0.08348317492936039,
"grad_norm": 0.21027198433876038,
"learning_rate": 1.1045284632676535e-05,
"loss": 0.1103,
"step": 1300
},
{
"epoch": 0.08669406627279733,
"grad_norm": 0.16708673536777496,
"learning_rate": 1.0348994967025012e-05,
"loss": 0.1085,
"step": 1350
},
{
"epoch": 0.08990495761623427,
"grad_norm": 0.1804681122303009,
"learning_rate": 9.651005032974994e-06,
"loss": 0.1104,
"step": 1400
},
{
"epoch": 0.0931158489596712,
"grad_norm": 0.19710887968540192,
"learning_rate": 8.954715367323468e-06,
"loss": 0.1102,
"step": 1450
},
{
"epoch": 0.09632674030310814,
"grad_norm": 0.17004291713237762,
"learning_rate": 8.263518223330698e-06,
"loss": 0.1065,
"step": 1500
},
{
"epoch": 0.09953763164654508,
"grad_norm": 0.184707373380661,
"learning_rate": 7.580781044003324e-06,
"loss": 0.1095,
"step": 1550
},
{
"epoch": 0.10274852298998202,
"grad_norm": 0.1840941607952118,
"learning_rate": 6.909830056250527e-06,
"loss": 0.1082,
"step": 1600
},
{
"epoch": 0.10595941433341896,
"grad_norm": 0.1997130811214447,
"learning_rate": 6.25393406584088e-06,
"loss": 0.1065,
"step": 1650
},
{
"epoch": 0.1091703056768559,
"grad_norm": 0.18672604858875275,
"learning_rate": 5.616288532109225e-06,
"loss": 0.1053,
"step": 1700
},
{
"epoch": 0.11238119702029284,
"grad_norm": 0.17353491485118866,
"learning_rate": 5.000000000000003e-06,
"loss": 0.1051,
"step": 1750
},
{
"epoch": 0.11559208836372978,
"grad_norm": 0.1928931623697281,
"learning_rate": 4.408070965292534e-06,
"loss": 0.1036,
"step": 1800
},
{
"epoch": 0.1188029797071667,
"grad_norm": 0.19479507207870483,
"learning_rate": 3.8433852467434175e-06,
"loss": 0.1014,
"step": 1850
},
{
"epoch": 0.12201387105060364,
"grad_norm": 0.17657093703746796,
"learning_rate": 3.308693936411421e-06,
"loss": 0.1036,
"step": 1900
},
{
"epoch": 0.1252247623940406,
"grad_norm": 0.17746590077877045,
"learning_rate": 2.8066019966134907e-06,
"loss": 0.1012,
"step": 1950
},
{
"epoch": 0.12843565373747753,
"grad_norm": 0.18518978357315063,
"learning_rate": 2.339555568810221e-06,
"loss": 0.106,
"step": 2000
}
],
"logging_steps": 50,
"max_steps": 2500,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 5.196245781970944e+18,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}