Nemo_12B_GENAI_1A / trainer_state.json
1024m's picture
Upload folder using huggingface_hub
7039742 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9997597886139803,
"eval_steps": 500,
"global_step": 2081,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.024021138601969732,
"grad_norm": 0.4454003870487213,
"learning_rate": 2e-05,
"loss": 2.2416,
"step": 50
},
{
"epoch": 0.048042277203939464,
"grad_norm": 0.5095623731613159,
"learning_rate": 1.9507631708517972e-05,
"loss": 1.7858,
"step": 100
},
{
"epoch": 0.0720634158059092,
"grad_norm": 0.2806667685508728,
"learning_rate": 1.9015263417035946e-05,
"loss": 1.6981,
"step": 150
},
{
"epoch": 0.09608455440787893,
"grad_norm": 0.28293704986572266,
"learning_rate": 1.8522895125553916e-05,
"loss": 1.7156,
"step": 200
},
{
"epoch": 0.12010569300984866,
"grad_norm": 0.31131088733673096,
"learning_rate": 1.8030526834071887e-05,
"loss": 1.7039,
"step": 250
},
{
"epoch": 0.1441268316118184,
"grad_norm": 0.34635475277900696,
"learning_rate": 1.7538158542589857e-05,
"loss": 1.7054,
"step": 300
},
{
"epoch": 0.16814797021378813,
"grad_norm": 0.4496716856956482,
"learning_rate": 1.704579025110783e-05,
"loss": 1.6978,
"step": 350
},
{
"epoch": 0.19216910881575786,
"grad_norm": 0.34168896079063416,
"learning_rate": 1.6553421959625802e-05,
"loss": 1.6664,
"step": 400
},
{
"epoch": 0.2161902474177276,
"grad_norm": 0.3718103766441345,
"learning_rate": 1.6061053668143772e-05,
"loss": 1.6731,
"step": 450
},
{
"epoch": 0.24021138601969733,
"grad_norm": 0.3723597228527069,
"learning_rate": 1.5568685376661746e-05,
"loss": 1.671,
"step": 500
},
{
"epoch": 0.2642325246216671,
"grad_norm": 0.3750188648700714,
"learning_rate": 1.5076317085179715e-05,
"loss": 1.6883,
"step": 550
},
{
"epoch": 0.2882536632236368,
"grad_norm": 0.35369986295700073,
"learning_rate": 1.4583948793697687e-05,
"loss": 1.6777,
"step": 600
},
{
"epoch": 0.3122748018256065,
"grad_norm": 0.4338513910770416,
"learning_rate": 1.4091580502215658e-05,
"loss": 1.6664,
"step": 650
},
{
"epoch": 0.33629594042757627,
"grad_norm": 0.36728423833847046,
"learning_rate": 1.359921221073363e-05,
"loss": 1.6343,
"step": 700
},
{
"epoch": 0.360317079029546,
"grad_norm": 0.4234764277935028,
"learning_rate": 1.3106843919251602e-05,
"loss": 1.7187,
"step": 750
},
{
"epoch": 0.3843382176315157,
"grad_norm": 0.41642165184020996,
"learning_rate": 1.2614475627769574e-05,
"loss": 1.7164,
"step": 800
},
{
"epoch": 0.40835935623348546,
"grad_norm": 0.42162206768989563,
"learning_rate": 1.2122107336287543e-05,
"loss": 1.7247,
"step": 850
},
{
"epoch": 0.4323804948354552,
"grad_norm": 0.42779281735420227,
"learning_rate": 1.1629739044805515e-05,
"loss": 1.644,
"step": 900
},
{
"epoch": 0.45640163343742496,
"grad_norm": 0.417507141828537,
"learning_rate": 1.1137370753323486e-05,
"loss": 1.6829,
"step": 950
},
{
"epoch": 0.48042277203939465,
"grad_norm": 0.6434239149093628,
"learning_rate": 1.0645002461841458e-05,
"loss": 1.6499,
"step": 1000
},
{
"epoch": 0.5044439106413644,
"grad_norm": 0.4976539611816406,
"learning_rate": 1.015263417035943e-05,
"loss": 1.6782,
"step": 1050
},
{
"epoch": 0.5284650492433342,
"grad_norm": 0.428740531206131,
"learning_rate": 9.6602658788774e-06,
"loss": 1.659,
"step": 1100
},
{
"epoch": 0.5524861878453039,
"grad_norm": 0.46981343626976013,
"learning_rate": 9.167897587395373e-06,
"loss": 1.6811,
"step": 1150
},
{
"epoch": 0.5765073264472736,
"grad_norm": 0.4911981225013733,
"learning_rate": 8.675529295913345e-06,
"loss": 1.6418,
"step": 1200
},
{
"epoch": 0.6005284650492433,
"grad_norm": 0.4720385670661926,
"learning_rate": 8.183161004431315e-06,
"loss": 1.6594,
"step": 1250
},
{
"epoch": 0.624549603651213,
"grad_norm": 0.7791111469268799,
"learning_rate": 7.690792712949286e-06,
"loss": 1.6814,
"step": 1300
},
{
"epoch": 0.6485707422531828,
"grad_norm": 0.5131087303161621,
"learning_rate": 7.198424421467258e-06,
"loss": 1.6509,
"step": 1350
},
{
"epoch": 0.6725918808551525,
"grad_norm": 0.4304710328578949,
"learning_rate": 6.70605612998523e-06,
"loss": 1.6751,
"step": 1400
},
{
"epoch": 0.6966130194571223,
"grad_norm": 0.4925595819950104,
"learning_rate": 6.213687838503201e-06,
"loss": 1.6418,
"step": 1450
},
{
"epoch": 0.720634158059092,
"grad_norm": 0.45314958691596985,
"learning_rate": 5.721319547021172e-06,
"loss": 1.6763,
"step": 1500
},
{
"epoch": 0.7446552966610618,
"grad_norm": 0.5018033981323242,
"learning_rate": 5.228951255539144e-06,
"loss": 1.6763,
"step": 1550
},
{
"epoch": 0.7686764352630314,
"grad_norm": 0.41523104906082153,
"learning_rate": 4.7365829640571156e-06,
"loss": 1.6727,
"step": 1600
},
{
"epoch": 0.7926975738650012,
"grad_norm": 0.4479306936264038,
"learning_rate": 4.244214672575087e-06,
"loss": 1.6588,
"step": 1650
},
{
"epoch": 0.8167187124669709,
"grad_norm": 0.504392147064209,
"learning_rate": 3.751846381093058e-06,
"loss": 1.6946,
"step": 1700
},
{
"epoch": 0.8407398510689407,
"grad_norm": 0.48174819350242615,
"learning_rate": 3.2594780896110296e-06,
"loss": 1.7111,
"step": 1750
},
{
"epoch": 0.8647609896709104,
"grad_norm": 0.6309037804603577,
"learning_rate": 2.767109798129001e-06,
"loss": 1.651,
"step": 1800
},
{
"epoch": 0.8887821282728802,
"grad_norm": 0.5154911279678345,
"learning_rate": 2.2747415066469722e-06,
"loss": 1.696,
"step": 1850
},
{
"epoch": 0.9128032668748499,
"grad_norm": 0.39096760749816895,
"learning_rate": 1.7823732151649436e-06,
"loss": 1.6385,
"step": 1900
},
{
"epoch": 0.9368244054768196,
"grad_norm": 0.4384596645832062,
"learning_rate": 1.290004923682915e-06,
"loss": 1.6523,
"step": 1950
},
{
"epoch": 0.9608455440787893,
"grad_norm": 0.48874229192733765,
"learning_rate": 7.976366322008864e-07,
"loss": 1.6501,
"step": 2000
},
{
"epoch": 0.984866682680759,
"grad_norm": 0.3902250826358795,
"learning_rate": 3.0526834071885774e-07,
"loss": 1.6395,
"step": 2050
}
],
"logging_steps": 50,
"max_steps": 2081,
"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": true
},
"attributes": {}
}
},
"total_flos": 1.5068479218407424e+18,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}