gpt2-xl-lora-multi-6 / trainer_state.json
MHGanainy's picture
MHGanainy/gpt2-xl-lora-multi-6
0f457ab verified
raw
history blame
11.1 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 5805,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.017226528854435832,
"grad_norm": 0.04875866696238518,
"learning_rate": 4.307559767391773e-07,
"loss": 2.816,
"step": 100
},
{
"epoch": 0.034453057708871665,
"grad_norm": 0.04928815737366676,
"learning_rate": 8.615119534783546e-07,
"loss": 2.8077,
"step": 200
},
{
"epoch": 0.05167958656330749,
"grad_norm": 0.05480952933430672,
"learning_rate": 1.292267930217532e-06,
"loss": 2.8162,
"step": 300
},
{
"epoch": 0.06890611541774333,
"grad_norm": 0.06494510173797607,
"learning_rate": 1.7230239069567093e-06,
"loss": 2.8237,
"step": 400
},
{
"epoch": 0.08613264427217916,
"grad_norm": 0.08336891233921051,
"learning_rate": 2.1537798836958863e-06,
"loss": 2.7987,
"step": 500
},
{
"epoch": 0.10335917312661498,
"grad_norm": 0.09097711741924286,
"learning_rate": 2.584535860435064e-06,
"loss": 2.7894,
"step": 600
},
{
"epoch": 0.12058570198105081,
"grad_norm": 0.11103380471467972,
"learning_rate": 3.015291837174241e-06,
"loss": 2.7909,
"step": 700
},
{
"epoch": 0.13781223083548666,
"grad_norm": 0.11578191071748734,
"learning_rate": 3.4460478139134186e-06,
"loss": 2.7922,
"step": 800
},
{
"epoch": 0.15503875968992248,
"grad_norm": 0.1395414024591446,
"learning_rate": 3.876803790652596e-06,
"loss": 2.7704,
"step": 900
},
{
"epoch": 0.17226528854435832,
"grad_norm": 0.1293153464794159,
"learning_rate": 4.3075597673917726e-06,
"loss": 2.7549,
"step": 1000
},
{
"epoch": 0.18949181739879414,
"grad_norm": 0.13815675675868988,
"learning_rate": 4.73831574413095e-06,
"loss": 2.7386,
"step": 1100
},
{
"epoch": 0.20671834625322996,
"grad_norm": 0.15647341310977936,
"learning_rate": 5.169071720870128e-06,
"loss": 2.7453,
"step": 1200
},
{
"epoch": 0.2239448751076658,
"grad_norm": 0.17875690758228302,
"learning_rate": 5.599827697609305e-06,
"loss": 2.7357,
"step": 1300
},
{
"epoch": 0.24117140396210163,
"grad_norm": 0.16741064190864563,
"learning_rate": 6.030583674348482e-06,
"loss": 2.739,
"step": 1400
},
{
"epoch": 0.25839793281653745,
"grad_norm": 0.17629371583461761,
"learning_rate": 6.461339651087659e-06,
"loss": 2.7257,
"step": 1500
},
{
"epoch": 0.2756244616709733,
"grad_norm": 0.18484149873256683,
"learning_rate": 6.892095627826837e-06,
"loss": 2.712,
"step": 1600
},
{
"epoch": 0.29285099052540914,
"grad_norm": 0.18326900899410248,
"learning_rate": 7.322851604566014e-06,
"loss": 2.7207,
"step": 1700
},
{
"epoch": 0.31007751937984496,
"grad_norm": 0.20282337069511414,
"learning_rate": 7.753607581305192e-06,
"loss": 2.7264,
"step": 1800
},
{
"epoch": 0.3273040482342808,
"grad_norm": 0.19705314934253693,
"learning_rate": 8.184363558044369e-06,
"loss": 2.7107,
"step": 1900
},
{
"epoch": 0.34453057708871665,
"grad_norm": 0.2224486768245697,
"learning_rate": 8.615119534783545e-06,
"loss": 2.6999,
"step": 2000
},
{
"epoch": 0.36175710594315247,
"grad_norm": 0.21177536249160767,
"learning_rate": 9.045875511522723e-06,
"loss": 2.7055,
"step": 2100
},
{
"epoch": 0.3789836347975883,
"grad_norm": 0.21247607469558716,
"learning_rate": 9.4766314882619e-06,
"loss": 2.6962,
"step": 2200
},
{
"epoch": 0.3962101636520241,
"grad_norm": 0.2235407829284668,
"learning_rate": 9.907387465001078e-06,
"loss": 2.6898,
"step": 2300
},
{
"epoch": 0.4134366925064599,
"grad_norm": 0.22821404039859772,
"learning_rate": 1.0338143441740256e-05,
"loss": 2.6896,
"step": 2400
},
{
"epoch": 0.4306632213608958,
"grad_norm": 0.21626809239387512,
"learning_rate": 1.0768899418479433e-05,
"loss": 2.6859,
"step": 2500
},
{
"epoch": 0.4478897502153316,
"grad_norm": 0.2285173088312149,
"learning_rate": 1.119965539521861e-05,
"loss": 2.6636,
"step": 2600
},
{
"epoch": 0.46511627906976744,
"grad_norm": 0.2364313304424286,
"learning_rate": 1.1630411371957786e-05,
"loss": 2.6726,
"step": 2700
},
{
"epoch": 0.48234280792420325,
"grad_norm": 0.2359386384487152,
"learning_rate": 1.2061167348696965e-05,
"loss": 2.6635,
"step": 2800
},
{
"epoch": 0.49956933677863913,
"grad_norm": 0.2400447577238083,
"learning_rate": 1.2491923325436141e-05,
"loss": 2.6934,
"step": 2900
},
{
"epoch": 0.5167958656330749,
"grad_norm": 0.24956579506397247,
"learning_rate": 1.2922679302175318e-05,
"loss": 2.6663,
"step": 3000
},
{
"epoch": 0.5340223944875108,
"grad_norm": 0.24709786474704742,
"learning_rate": 1.3353435278914494e-05,
"loss": 2.6408,
"step": 3100
},
{
"epoch": 0.5512489233419466,
"grad_norm": 0.2572828233242035,
"learning_rate": 1.3784191255653674e-05,
"loss": 2.6621,
"step": 3200
},
{
"epoch": 0.5684754521963824,
"grad_norm": 0.26790568232536316,
"learning_rate": 1.421494723239285e-05,
"loss": 2.6388,
"step": 3300
},
{
"epoch": 0.5857019810508183,
"grad_norm": 0.2591148912906647,
"learning_rate": 1.4645703209132027e-05,
"loss": 2.6415,
"step": 3400
},
{
"epoch": 0.602928509905254,
"grad_norm": 0.2609754800796509,
"learning_rate": 1.5076459185871206e-05,
"loss": 2.6588,
"step": 3500
},
{
"epoch": 0.6201550387596899,
"grad_norm": 0.2639792859554291,
"learning_rate": 1.5507215162610384e-05,
"loss": 2.6533,
"step": 3600
},
{
"epoch": 0.6373815676141258,
"grad_norm": 0.2771030366420746,
"learning_rate": 1.593797113934956e-05,
"loss": 2.6316,
"step": 3700
},
{
"epoch": 0.6546080964685616,
"grad_norm": 0.2596473693847656,
"learning_rate": 1.6368727116088737e-05,
"loss": 2.6474,
"step": 3800
},
{
"epoch": 0.6718346253229974,
"grad_norm": 0.2842521369457245,
"learning_rate": 1.6799483092827914e-05,
"loss": 2.6475,
"step": 3900
},
{
"epoch": 0.6890611541774333,
"grad_norm": 0.282993346452713,
"learning_rate": 1.723023906956709e-05,
"loss": 2.6391,
"step": 4000
},
{
"epoch": 0.7062876830318691,
"grad_norm": 0.29125678539276123,
"learning_rate": 1.766099504630627e-05,
"loss": 2.6246,
"step": 4100
},
{
"epoch": 0.7235142118863049,
"grad_norm": 0.2812328338623047,
"learning_rate": 1.8091751023045447e-05,
"loss": 2.6469,
"step": 4200
},
{
"epoch": 0.7407407407407407,
"grad_norm": 0.2858037054538727,
"learning_rate": 1.8522506999784623e-05,
"loss": 2.6348,
"step": 4300
},
{
"epoch": 0.7579672695951766,
"grad_norm": 0.30100327730178833,
"learning_rate": 1.89532629765238e-05,
"loss": 2.6315,
"step": 4400
},
{
"epoch": 0.7751937984496124,
"grad_norm": 0.28023988008499146,
"learning_rate": 1.9384018953262976e-05,
"loss": 2.6244,
"step": 4500
},
{
"epoch": 0.7924203273040482,
"grad_norm": 0.2947770953178406,
"learning_rate": 1.9814774930002156e-05,
"loss": 2.6277,
"step": 4600
},
{
"epoch": 0.8096468561584841,
"grad_norm": 0.30242860317230225,
"learning_rate": 1.9881492033537068e-05,
"loss": 2.6132,
"step": 4700
},
{
"epoch": 0.8268733850129198,
"grad_norm": 0.2965434789657593,
"learning_rate": 1.9112585857815527e-05,
"loss": 2.6114,
"step": 4800
},
{
"epoch": 0.8440999138673557,
"grad_norm": 0.2951013743877411,
"learning_rate": 1.7681642907588395e-05,
"loss": 2.6349,
"step": 4900
},
{
"epoch": 0.8613264427217916,
"grad_norm": 0.34424757957458496,
"learning_rate": 1.5692622352080662e-05,
"loss": 2.6065,
"step": 5000
},
{
"epoch": 0.8785529715762274,
"grad_norm": 0.283710241317749,
"learning_rate": 1.329002815246645e-05,
"loss": 2.6098,
"step": 5100
},
{
"epoch": 0.8957795004306632,
"grad_norm": 0.28349170088768005,
"learning_rate": 1.064841073155404e-05,
"loss": 2.618,
"step": 5200
},
{
"epoch": 0.9130060292850991,
"grad_norm": 0.31287553906440735,
"learning_rate": 7.95968574359416e-06,
"loss": 2.603,
"step": 5300
},
{
"epoch": 0.9302325581395349,
"grad_norm": 0.2843669652938843,
"learning_rate": 5.419191245899586e-06,
"loss": 2.6192,
"step": 5400
},
{
"epoch": 0.9474590869939707,
"grad_norm": 0.2816653847694397,
"learning_rate": 3.211496230098576e-06,
"loss": 2.6142,
"step": 5500
},
{
"epoch": 0.9646856158484065,
"grad_norm": 0.30356064438819885,
"learning_rate": 1.496991534769564e-06,
"loss": 2.6198,
"step": 5600
},
{
"epoch": 0.9819121447028424,
"grad_norm": 0.27873966097831726,
"learning_rate": 4.0023732056077235e-07,
"loss": 2.6062,
"step": 5700
},
{
"epoch": 0.9991386735572783,
"grad_norm": 0.3346691429615021,
"learning_rate": 9.136732052228426e-10,
"loss": 2.6072,
"step": 5800
},
{
"epoch": 1.0,
"step": 5805,
"total_flos": 8.45892909419987e+17,
"train_loss": 2.683517217184325,
"train_runtime": 1637.9851,
"train_samples_per_second": 56.701,
"train_steps_per_second": 3.544
}
],
"logging_steps": 100,
"max_steps": 5805,
"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": 8.45892909419987e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}