Llama-3-lora / trainer_state.json
xi0v's picture
Upload folder using huggingface_hub
ec9bb8b verified
raw
history blame contribute delete
No virus
7.94 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.989010989010989,
"eval_steps": 500,
"global_step": 408,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07326007326007326,
"grad_norm": 2.0240988731384277,
"learning_rate": 1.2195121951219513e-05,
"loss": 1.493,
"step": 10
},
{
"epoch": 0.14652014652014653,
"grad_norm": 1.0480456352233887,
"learning_rate": 2.4390243902439026e-05,
"loss": 1.1235,
"step": 20
},
{
"epoch": 0.21978021978021978,
"grad_norm": 0.3642168939113617,
"learning_rate": 3.6585365853658535e-05,
"loss": 1.1219,
"step": 30
},
{
"epoch": 0.29304029304029305,
"grad_norm": 0.4966081380844116,
"learning_rate": 4.878048780487805e-05,
"loss": 1.1664,
"step": 40
},
{
"epoch": 0.3663003663003663,
"grad_norm": 0.42003196477890015,
"learning_rate": 4.992584373844853e-05,
"loss": 1.0602,
"step": 50
},
{
"epoch": 0.43956043956043955,
"grad_norm": 0.6992456316947937,
"learning_rate": 4.9670065861258264e-05,
"loss": 1.0122,
"step": 60
},
{
"epoch": 0.5128205128205128,
"grad_norm": 1.130461573600769,
"learning_rate": 4.9233623530708587e-05,
"loss": 1.1042,
"step": 70
},
{
"epoch": 0.5860805860805861,
"grad_norm": 1.0719728469848633,
"learning_rate": 4.861971291236772e-05,
"loss": 0.9888,
"step": 80
},
{
"epoch": 0.6593406593406593,
"grad_norm": 0.9262090921401978,
"learning_rate": 4.7832829812079794e-05,
"loss": 1.0949,
"step": 90
},
{
"epoch": 0.7326007326007326,
"grad_norm": 0.842754602432251,
"learning_rate": 4.687873675216522e-05,
"loss": 1.0847,
"step": 100
},
{
"epoch": 0.8058608058608059,
"grad_norm": 0.7069523930549622,
"learning_rate": 4.576442077117073e-05,
"loss": 1.0386,
"step": 110
},
{
"epoch": 0.8791208791208791,
"grad_norm": 1.297459602355957,
"learning_rate": 4.4498042256211164e-05,
"loss": 1.0071,
"step": 120
},
{
"epoch": 0.9523809523809523,
"grad_norm": 3.424927234649658,
"learning_rate": 4.308887518261507e-05,
"loss": 0.9977,
"step": 130
},
{
"epoch": 1.0256410256410255,
"grad_norm": 1.1368341445922852,
"learning_rate": 4.154723919851291e-05,
"loss": 1.0333,
"step": 140
},
{
"epoch": 1.098901098901099,
"grad_norm": 1.1726893186569214,
"learning_rate": 3.9884424051727556e-05,
"loss": 0.8241,
"step": 150
},
{
"epoch": 1.1721611721611722,
"grad_norm": 0.8981992602348328,
"learning_rate": 3.8112606912406044e-05,
"loss": 0.7771,
"step": 160
},
{
"epoch": 1.2454212454212454,
"grad_norm": 1.5086628198623657,
"learning_rate": 3.624476319685771e-05,
"loss": 0.7341,
"step": 170
},
{
"epoch": 1.3186813186813187,
"grad_norm": 0.9521022439002991,
"learning_rate": 3.429457154565565e-05,
"loss": 0.7059,
"step": 180
},
{
"epoch": 1.3919413919413919,
"grad_norm": 2.737773895263672,
"learning_rate": 3.227631365186836e-05,
"loss": 0.7403,
"step": 190
},
{
"epoch": 1.4652014652014653,
"grad_norm": 0.8319964408874512,
"learning_rate": 3.020476967300212e-05,
"loss": 0.7738,
"step": 200
},
{
"epoch": 1.5384615384615383,
"grad_norm": 1.0483458042144775,
"learning_rate": 2.8095109992575823e-05,
"loss": 0.6995,
"step": 210
},
{
"epoch": 1.6117216117216118,
"grad_norm": 1.0540846586227417,
"learning_rate": 2.5962784123982843e-05,
"loss": 0.7363,
"step": 220
},
{
"epoch": 1.684981684981685,
"grad_norm": 1.6841723918914795,
"learning_rate": 2.3823407570221813e-05,
"loss": 0.7735,
"step": 230
},
{
"epoch": 1.7582417582417582,
"grad_norm": 1.5593695640563965,
"learning_rate": 2.1692647468048236e-05,
"loss": 0.6975,
"step": 240
},
{
"epoch": 1.8315018315018317,
"grad_norm": 0.8219186067581177,
"learning_rate": 1.9586107854000326e-05,
"loss": 0.7239,
"step": 250
},
{
"epoch": 1.9047619047619047,
"grad_norm": 1.9310203790664673,
"learning_rate": 1.7519215392522025e-05,
"loss": 0.7285,
"step": 260
},
{
"epoch": 1.978021978021978,
"grad_norm": 0.888729989528656,
"learning_rate": 1.5507106403021894e-05,
"loss": 0.7415,
"step": 270
},
{
"epoch": 2.051282051282051,
"grad_norm": 0.5562293529510498,
"learning_rate": 1.3564516013194023e-05,
"loss": 0.5019,
"step": 280
},
{
"epoch": 2.1245421245421245,
"grad_norm": 1.5220221281051636,
"learning_rate": 1.1705670250356415e-05,
"loss": 0.5581,
"step": 290
},
{
"epoch": 2.197802197802198,
"grad_norm": 1.251187801361084,
"learning_rate": 9.944181861046186e-06,
"loss": 0.495,
"step": 300
},
{
"epoch": 2.271062271062271,
"grad_norm": 0.9048457145690918,
"learning_rate": 8.292950621808022e-06,
"loss": 0.5017,
"step": 310
},
{
"epoch": 2.3443223443223444,
"grad_norm": 0.9046669602394104,
"learning_rate": 6.764068871222826e-06,
"loss": 0.5482,
"step": 320
},
{
"epoch": 2.4175824175824174,
"grad_norm": 1.1838064193725586,
"learning_rate": 5.368732954986388e-06,
"loss": 0.4744,
"step": 330
},
{
"epoch": 2.490842490842491,
"grad_norm": 0.6814486384391785,
"learning_rate": 4.117161232546024e-06,
"loss": 0.5586,
"step": 340
},
{
"epoch": 2.564102564102564,
"grad_norm": 0.7199404239654541,
"learning_rate": 3.0185192457509894e-06,
"loss": 0.4924,
"step": 350
},
{
"epoch": 2.6373626373626373,
"grad_norm": 1.0055203437805176,
"learning_rate": 2.0808525975233805e-06,
"loss": 0.4985,
"step": 360
},
{
"epoch": 2.7106227106227108,
"grad_norm": 0.7133710384368896,
"learning_rate": 1.311028032094369e-06,
"loss": 0.4771,
"step": 370
},
{
"epoch": 2.7838827838827838,
"grad_norm": 1.280009388923645,
"learning_rate": 7.146831482883115e-07,
"loss": 0.4619,
"step": 380
},
{
"epoch": 2.857142857142857,
"grad_norm": 1.0090081691741943,
"learning_rate": 2.961851141157046e-07,
"loss": 0.5103,
"step": 390
},
{
"epoch": 2.9304029304029307,
"grad_norm": 0.7528389096260071,
"learning_rate": 5.85986850174608e-08,
"loss": 0.5113,
"step": 400
},
{
"epoch": 2.989010989010989,
"step": 408,
"total_flos": 3.5468284402335744e+16,
"train_loss": 0.7816439504716911,
"train_runtime": 3976.7208,
"train_samples_per_second": 0.823,
"train_steps_per_second": 0.103
}
],
"logging_steps": 10,
"max_steps": 408,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 1000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.5468284402335744e+16,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}