gpt2_f_experiment_drug_data_large / trainer_state.json
mllm-dev's picture
Upload folder using huggingface_hub
5897502 verified
{
"best_metric": 1.4546449184417725,
"best_model_checkpoint": "tam_test_out_drug_data_large/checkpoint-13442",
"epoch": 1.0,
"eval_steps": 500,
"global_step": 13442,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"grad_norm": 364818.0,
"learning_rate": 5.7768189257550956e-05,
"loss": 1.9818,
"step": 500
},
{
"epoch": 0.07,
"grad_norm": 400334.53125,
"learning_rate": 5.5536378515101923e-05,
"loss": 1.7592,
"step": 1000
},
{
"epoch": 0.11,
"grad_norm": 406113.28125,
"learning_rate": 5.330456777265288e-05,
"loss": 1.7118,
"step": 1500
},
{
"epoch": 0.15,
"grad_norm": 331607.125,
"learning_rate": 5.1072757030203845e-05,
"loss": 1.7036,
"step": 2000
},
{
"epoch": 0.19,
"grad_norm": 390128.875,
"learning_rate": 4.88409462877548e-05,
"loss": 1.6583,
"step": 2500
},
{
"epoch": 0.22,
"grad_norm": 339776.03125,
"learning_rate": 4.6609135545305754e-05,
"loss": 1.6358,
"step": 3000
},
{
"epoch": 0.26,
"grad_norm": 782641.125,
"learning_rate": 4.437732480285672e-05,
"loss": 1.5937,
"step": 3500
},
{
"epoch": 0.3,
"grad_norm": 600757.25,
"learning_rate": 4.2145514060407676e-05,
"loss": 1.6129,
"step": 4000
},
{
"epoch": 0.33,
"grad_norm": 275153.03125,
"learning_rate": 3.9913703317958644e-05,
"loss": 1.6207,
"step": 4500
},
{
"epoch": 0.37,
"grad_norm": 471402.28125,
"learning_rate": 3.76818925755096e-05,
"loss": 1.5781,
"step": 5000
},
{
"epoch": 0.41,
"grad_norm": 225413.765625,
"learning_rate": 3.545008183306055e-05,
"loss": 1.576,
"step": 5500
},
{
"epoch": 0.45,
"grad_norm": 553825.5,
"learning_rate": 3.321827109061152e-05,
"loss": 1.5546,
"step": 6000
},
{
"epoch": 0.48,
"grad_norm": 228419.859375,
"learning_rate": 3.0986460348162474e-05,
"loss": 1.5777,
"step": 6500
},
{
"epoch": 0.52,
"grad_norm": 461608.46875,
"learning_rate": 2.8754649605713435e-05,
"loss": 1.5543,
"step": 7000
},
{
"epoch": 0.56,
"grad_norm": 303856.875,
"learning_rate": 2.6522838863264396e-05,
"loss": 1.5475,
"step": 7500
},
{
"epoch": 0.6,
"grad_norm": 366325.34375,
"learning_rate": 2.4291028120815357e-05,
"loss": 1.5295,
"step": 8000
},
{
"epoch": 0.63,
"grad_norm": 407897.65625,
"learning_rate": 2.2059217378366318e-05,
"loss": 1.5193,
"step": 8500
},
{
"epoch": 0.67,
"grad_norm": 359063.25,
"learning_rate": 1.9827406635917272e-05,
"loss": 1.5173,
"step": 9000
},
{
"epoch": 0.71,
"grad_norm": 632423.8125,
"learning_rate": 1.7595595893468233e-05,
"loss": 1.497,
"step": 9500
},
{
"epoch": 0.74,
"grad_norm": 428668.46875,
"learning_rate": 1.5363785151019194e-05,
"loss": 1.5284,
"step": 10000
},
{
"epoch": 0.78,
"grad_norm": 406600.40625,
"learning_rate": 1.3131974408570155e-05,
"loss": 1.4979,
"step": 10500
},
{
"epoch": 0.82,
"grad_norm": 353293.84375,
"learning_rate": 1.0900163666121113e-05,
"loss": 1.4957,
"step": 11000
},
{
"epoch": 0.86,
"grad_norm": 281095.96875,
"learning_rate": 8.668352923672074e-06,
"loss": 1.4742,
"step": 11500
},
{
"epoch": 0.89,
"grad_norm": 332004.0625,
"learning_rate": 6.436542181223033e-06,
"loss": 1.4797,
"step": 12000
},
{
"epoch": 0.93,
"grad_norm": 299908.8125,
"learning_rate": 4.204731438773992e-06,
"loss": 1.47,
"step": 12500
},
{
"epoch": 0.97,
"grad_norm": 487343.84375,
"learning_rate": 1.972920696324952e-06,
"loss": 1.4685,
"step": 13000
},
{
"epoch": 1.0,
"eval_accuracy": 0.4630335720264112,
"eval_loss": 1.4546449184417725,
"eval_runtime": 302.6957,
"eval_samples_per_second": 35.524,
"eval_steps_per_second": 2.963,
"step": 13442
}
],
"logging_steps": 500,
"max_steps": 13442,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"total_flos": 2.945281604340096e+17,
"train_batch_size": 12,
"trial_name": null,
"trial_params": null
}