kizaru / trainer_state.json
AlekseyKorshuk's picture
huggingartists
71ecf5a
{
"best_metric": 1.865520715713501,
"best_model_checkpoint": "output/kizaru/checkpoint-146",
"epoch": 1.0,
"global_step": 146,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.0001368033484129251,
"loss": 2.5197,
"step": 5
},
{
"epoch": 0.07,
"learning_rate": 0.00013561798060451468,
"loss": 2.3915,
"step": 10
},
{
"epoch": 0.1,
"learning_rate": 0.00013365760438883496,
"loss": 2.3406,
"step": 15
},
{
"epoch": 0.14,
"learning_rate": 0.0001309448899214899,
"loss": 2.1122,
"step": 20
},
{
"epoch": 0.17,
"learning_rate": 0.0001275112075377201,
"loss": 2.2383,
"step": 25
},
{
"epoch": 0.21,
"learning_rate": 0.00012339626498000464,
"loss": 2.0867,
"step": 30
},
{
"epoch": 0.24,
"learning_rate": 0.00011864764821033304,
"loss": 2.1937,
"step": 35
},
{
"epoch": 0.27,
"learning_rate": 0.00011332027111727967,
"loss": 2.0148,
"step": 40
},
{
"epoch": 0.31,
"learning_rate": 0.00010747574048157067,
"loss": 2.0935,
"step": 45
},
{
"epoch": 0.34,
"learning_rate": 0.00010118164354379966,
"loss": 2.0352,
"step": 50
},
{
"epoch": 0.38,
"learning_rate": 9.451076641299198e-05,
"loss": 2.0251,
"step": 55
},
{
"epoch": 0.41,
"learning_rate": 8.754025235448638e-05,
"loss": 2.0361,
"step": 60
},
{
"epoch": 0.45,
"learning_rate": 8.035070969084936e-05,
"loss": 1.9591,
"step": 65
},
{
"epoch": 0.48,
"learning_rate": 7.302527963222311e-05,
"loss": 1.9692,
"step": 70
},
{
"epoch": 0.51,
"learning_rate": 6.56486748158892e-05,
"loss": 2.2105,
"step": 75
},
{
"epoch": 0.55,
"learning_rate": 5.830619967355683e-05,
"loss": 2.0882,
"step": 80
},
{
"epoch": 0.58,
"learning_rate": 5.108276395502957e-05,
"loss": 1.9231,
"step": 85
},
{
"epoch": 0.62,
"learning_rate": 4.406190081604759e-05,
"loss": 1.9177,
"step": 90
},
{
"epoch": 0.65,
"learning_rate": 3.7324800825319744e-05,
"loss": 1.7732,
"step": 95
},
{
"epoch": 0.68,
"learning_rate": 3.094937306166828e-05,
"loss": 1.9072,
"step": 100
},
{
"epoch": 0.72,
"learning_rate": 2.5009344158933835e-05,
"loss": 1.8601,
"step": 105
},
{
"epoch": 0.75,
"learning_rate": 1.957340571745288e-05,
"loss": 1.9745,
"step": 110
},
{
"epoch": 0.79,
"learning_rate": 1.4704419941600183e-05,
"loss": 2.0633,
"step": 115
},
{
"epoch": 0.82,
"learning_rate": 1.0458692689551801e-05,
"loss": 2.0157,
"step": 120
},
{
"epoch": 0.86,
"learning_rate": 6.885322341856113e-06,
"loss": 1.8486,
"step": 125
},
{
"epoch": 0.89,
"learning_rate": 4.02563201861869e-06,
"loss": 1.8927,
"step": 130
},
{
"epoch": 0.92,
"learning_rate": 1.9126917112463875e-06,
"loss": 1.8729,
"step": 135
},
{
"epoch": 0.96,
"learning_rate": 5.709358549087076e-07,
"loss": 1.85,
"step": 140
},
{
"epoch": 0.99,
"learning_rate": 1.588076417948654e-08,
"loss": 1.964,
"step": 145
},
{
"epoch": 1.0,
"eval_loss": 1.865520715713501,
"eval_runtime": 8.7722,
"eval_samples_per_second": 21.089,
"eval_steps_per_second": 2.736,
"step": 146
}
],
"max_steps": 146,
"num_train_epochs": 1,
"total_flos": 152071962624000.0,
"trial_name": null,
"trial_params": null
}