uniform-complex-gpt2-med / trainer_state.json
josephimperial's picture
Upload 14 files
2f20b34
raw
history blame
4.85 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.31358073359972766,
"global_step": 17500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.985067584114299e-05,
"loss": 2.6076,
"step": 500
},
{
"epoch": 0.02,
"learning_rate": 4.970135168228597e-05,
"loss": 2.5834,
"step": 1000
},
{
"epoch": 0.03,
"learning_rate": 4.955202752342896e-05,
"loss": 2.5955,
"step": 1500
},
{
"epoch": 0.04,
"learning_rate": 4.940270336457195e-05,
"loss": 2.5908,
"step": 2000
},
{
"epoch": 0.04,
"learning_rate": 4.925337920571494e-05,
"loss": 2.5974,
"step": 2500
},
{
"epoch": 0.05,
"learning_rate": 4.9104055046857924e-05,
"loss": 3.0537,
"step": 3000
},
{
"epoch": 0.06,
"learning_rate": 4.895473088800091e-05,
"loss": 3.0324,
"step": 3500
},
{
"epoch": 0.07,
"learning_rate": 4.88054067291439e-05,
"loss": 3.0167,
"step": 4000
},
{
"epoch": 0.08,
"learning_rate": 4.865608257028688e-05,
"loss": 3.0234,
"step": 4500
},
{
"epoch": 0.09,
"learning_rate": 4.850675841142987e-05,
"loss": 3.0113,
"step": 5000
},
{
"epoch": 0.1,
"learning_rate": 4.835743425257286e-05,
"loss": 3.0151,
"step": 5500
},
{
"epoch": 0.11,
"learning_rate": 4.8208110093715845e-05,
"loss": 2.9967,
"step": 6000
},
{
"epoch": 0.12,
"learning_rate": 4.805878593485883e-05,
"loss": 2.998,
"step": 6500
},
{
"epoch": 0.13,
"learning_rate": 4.790946177600182e-05,
"loss": 2.9781,
"step": 7000
},
{
"epoch": 0.13,
"learning_rate": 4.776013761714481e-05,
"loss": 2.9997,
"step": 7500
},
{
"epoch": 0.14,
"learning_rate": 4.761081345828779e-05,
"loss": 2.9812,
"step": 8000
},
{
"epoch": 0.15,
"learning_rate": 4.746148929943078e-05,
"loss": 2.97,
"step": 8500
},
{
"epoch": 0.16,
"learning_rate": 4.7312165140573766e-05,
"loss": 2.9958,
"step": 9000
},
{
"epoch": 0.17,
"learning_rate": 4.7162840981716754e-05,
"loss": 2.9818,
"step": 9500
},
{
"epoch": 0.18,
"learning_rate": 4.7013516822859735e-05,
"loss": 2.973,
"step": 10000
},
{
"epoch": 0.19,
"learning_rate": 4.686419266400272e-05,
"loss": 2.9652,
"step": 10500
},
{
"epoch": 0.2,
"learning_rate": 4.671486850514572e-05,
"loss": 2.9503,
"step": 11000
},
{
"epoch": 0.21,
"learning_rate": 4.65655443462887e-05,
"loss": 2.9483,
"step": 11500
},
{
"epoch": 0.22,
"learning_rate": 4.641622018743169e-05,
"loss": 2.9592,
"step": 12000
},
{
"epoch": 0.22,
"learning_rate": 4.6266896028574675e-05,
"loss": 2.9615,
"step": 12500
},
{
"epoch": 0.23,
"learning_rate": 4.611757186971766e-05,
"loss": 2.9652,
"step": 13000
},
{
"epoch": 0.24,
"learning_rate": 4.5968247710860644e-05,
"loss": 2.9406,
"step": 13500
},
{
"epoch": 0.25,
"learning_rate": 4.581892355200363e-05,
"loss": 2.9328,
"step": 14000
},
{
"epoch": 0.26,
"learning_rate": 4.566959939314662e-05,
"loss": 2.939,
"step": 14500
},
{
"epoch": 0.27,
"learning_rate": 4.552027523428961e-05,
"loss": 2.9252,
"step": 15000
},
{
"epoch": 0.28,
"learning_rate": 4.5370951075432596e-05,
"loss": 2.9352,
"step": 15500
},
{
"epoch": 0.29,
"learning_rate": 4.5221626916575584e-05,
"loss": 2.9564,
"step": 16000
},
{
"epoch": 0.3,
"learning_rate": 4.5072302757718565e-05,
"loss": 2.943,
"step": 16500
},
{
"epoch": 0.3,
"learning_rate": 4.492297859886155e-05,
"loss": 2.9285,
"step": 17000
},
{
"epoch": 0.31,
"learning_rate": 4.477365444000454e-05,
"loss": 2.929,
"step": 17500
},
{
"epoch": 0.31,
"step": 17500,
"total_flos": 3.250452430848e+16,
"train_loss": 0.0,
"train_runtime": 72.3486,
"train_samples_per_second": 505.635,
"train_steps_per_second": 63.235
}
],
"max_steps": 4575,
"num_train_epochs": 3,
"total_flos": 3.250452430848e+16,
"trial_name": null,
"trial_params": null
}