finetuned-websites / trainer_state.json
sksatyam's picture
🍻 cheers
41fc1b4 verified
{
"best_metric": 0.8348819613456726,
"best_model_checkpoint": "finetuned-websites/checkpoint-200",
"epoch": 10.0,
"eval_steps": 100,
"global_step": 240,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.4166666666666667,
"grad_norm": 6.133713722229004,
"learning_rate": 1e-05,
"loss": 1.0596,
"step": 10
},
{
"epoch": 0.8333333333333334,
"grad_norm": 5.39012336730957,
"learning_rate": 2e-05,
"loss": 1.0523,
"step": 20
},
{
"epoch": 1.25,
"grad_norm": 4.8409881591796875,
"learning_rate": 3e-05,
"loss": 1.02,
"step": 30
},
{
"epoch": 1.6666666666666665,
"grad_norm": 5.128672122955322,
"learning_rate": 4e-05,
"loss": 0.9759,
"step": 40
},
{
"epoch": 2.0833333333333335,
"grad_norm": 6.329381465911865,
"learning_rate": 5e-05,
"loss": 0.9153,
"step": 50
},
{
"epoch": 2.5,
"grad_norm": 6.218069076538086,
"learning_rate": 6e-05,
"loss": 0.8539,
"step": 60
},
{
"epoch": 2.9166666666666665,
"grad_norm": 5.5271430015563965,
"learning_rate": 7.000000000000001e-05,
"loss": 0.8747,
"step": 70
},
{
"epoch": 3.3333333333333335,
"grad_norm": 8.148289680480957,
"learning_rate": 8e-05,
"loss": 0.7716,
"step": 80
},
{
"epoch": 3.75,
"grad_norm": 5.050111293792725,
"learning_rate": 8.9e-05,
"loss": 0.8763,
"step": 90
},
{
"epoch": 4.166666666666667,
"grad_norm": 5.70500373840332,
"learning_rate": 9.900000000000001e-05,
"loss": 0.6802,
"step": 100
},
{
"epoch": 4.166666666666667,
"eval_accuracy": 0.5735294117647058,
"eval_loss": 1.2561215162277222,
"eval_runtime": 2.1458,
"eval_samples_per_second": 31.69,
"eval_steps_per_second": 4.194,
"step": 100
},
{
"epoch": 4.583333333333333,
"grad_norm": 6.188072204589844,
"learning_rate": 0.000109,
"loss": 0.6251,
"step": 110
},
{
"epoch": 5.0,
"grad_norm": 9.144834518432617,
"learning_rate": 0.00011899999999999999,
"loss": 0.6689,
"step": 120
},
{
"epoch": 5.416666666666667,
"grad_norm": 5.840977668762207,
"learning_rate": 0.00012900000000000002,
"loss": 0.5262,
"step": 130
},
{
"epoch": 5.833333333333333,
"grad_norm": 22.520524978637695,
"learning_rate": 0.00013900000000000002,
"loss": 0.6559,
"step": 140
},
{
"epoch": 6.25,
"grad_norm": 7.3676676750183105,
"learning_rate": 0.000149,
"loss": 0.5186,
"step": 150
},
{
"epoch": 6.666666666666667,
"grad_norm": 16.96535873413086,
"learning_rate": 0.00015900000000000002,
"loss": 0.4291,
"step": 160
},
{
"epoch": 7.083333333333333,
"grad_norm": 4.878167629241943,
"learning_rate": 0.00016900000000000002,
"loss": 0.6311,
"step": 170
},
{
"epoch": 7.5,
"grad_norm": 5.731603145599365,
"learning_rate": 0.000179,
"loss": 0.4274,
"step": 180
},
{
"epoch": 7.916666666666667,
"grad_norm": 12.268874168395996,
"learning_rate": 0.000189,
"loss": 0.3916,
"step": 190
},
{
"epoch": 8.333333333333334,
"grad_norm": 19.5472469329834,
"learning_rate": 0.000199,
"loss": 0.3727,
"step": 200
},
{
"epoch": 8.333333333333334,
"eval_accuracy": 0.75,
"eval_loss": 0.8348819613456726,
"eval_runtime": 2.0932,
"eval_samples_per_second": 32.486,
"eval_steps_per_second": 4.3,
"step": 200
},
{
"epoch": 8.75,
"grad_norm": 9.300850868225098,
"learning_rate": 0.00020899999999999998,
"loss": 0.5022,
"step": 210
},
{
"epoch": 9.166666666666666,
"grad_norm": 2.52441668510437,
"learning_rate": 0.000219,
"loss": 0.3873,
"step": 220
},
{
"epoch": 9.583333333333334,
"grad_norm": 7.8756794929504395,
"learning_rate": 0.000229,
"loss": 0.5663,
"step": 230
},
{
"epoch": 10.0,
"grad_norm": 7.985933780670166,
"learning_rate": 0.00023899999999999998,
"loss": 0.6521,
"step": 240
},
{
"epoch": 10.0,
"step": 240,
"total_flos": 2.9603266596900864e+17,
"train_loss": 0.6847666402657827,
"train_runtime": 183.3512,
"train_samples_per_second": 20.834,
"train_steps_per_second": 1.309
}
],
"logging_steps": 10,
"max_steps": 240,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.9603266596900864e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}