{
"best_metric": 0.08177243918180466,
"best_model_checkpoint": "./beans_outputs/checkpoint-520",
"epoch": 5.0,
"global_step": 650,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.08,
"learning_rate": 1.9692307692307696e-05,
"loss": 1.0383,
"step": 10
},
{
"epoch": 0.15,
"learning_rate": 1.9384615384615386e-05,
"loss": 0.9152,
"step": 20
},
{
"epoch": 0.23,
"learning_rate": 1.907692307692308e-05,
"loss": 0.8683,
"step": 30
},
{
"epoch": 0.31,
"learning_rate": 1.876923076923077e-05,
"loss": 0.7384,
"step": 40
},
{
"epoch": 0.38,
"learning_rate": 1.8461538461538465e-05,
"loss": 0.6344,
"step": 50
},
{
"epoch": 0.46,
"learning_rate": 1.8153846153846155e-05,
"loss": 0.6242,
"step": 60
},
{
"epoch": 0.54,
"learning_rate": 1.784615384615385e-05,
"loss": 0.5253,
"step": 70
},
{
"epoch": 0.62,
"learning_rate": 1.753846153846154e-05,
"loss": 0.5049,
"step": 80
},
{
"epoch": 0.69,
"learning_rate": 1.7230769230769234e-05,
"loss": 0.4103,
"step": 90
},
{
"epoch": 0.77,
"learning_rate": 1.6923076923076924e-05,
"loss": 0.37,
"step": 100
},
{
"epoch": 0.85,
"learning_rate": 1.6615384615384618e-05,
"loss": 0.3095,
"step": 110
},
{
"epoch": 0.92,
"learning_rate": 1.630769230769231e-05,
"loss": 0.3019,
"step": 120
},
{
"epoch": 1.0,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.254,
"step": 130
},
{
"epoch": 1.0,
"eval_accuracy": 0.9624060150375939,
"eval_loss": 0.22443579137325287,
"eval_runtime": 1.8627,
"eval_samples_per_second": 71.403,
"eval_steps_per_second": 9.127,
"step": 130
},
{
"epoch": 1.08,
"learning_rate": 1.5692307692307693e-05,
"loss": 0.2623,
"step": 140
},
{
"epoch": 1.15,
"learning_rate": 1.5384615384615387e-05,
"loss": 0.2771,
"step": 150
},
{
"epoch": 1.23,
"learning_rate": 1.5076923076923078e-05,
"loss": 0.2007,
"step": 160
},
{
"epoch": 1.31,
"learning_rate": 1.4769230769230772e-05,
"loss": 0.2666,
"step": 170
},
{
"epoch": 1.38,
"learning_rate": 1.4461538461538462e-05,
"loss": 0.2311,
"step": 180
},
{
"epoch": 1.46,
"learning_rate": 1.4153846153846156e-05,
"loss": 0.19,
"step": 190
},
{
"epoch": 1.54,
"learning_rate": 1.3846153846153847e-05,
"loss": 0.1593,
"step": 200
},
{
"epoch": 1.62,
"learning_rate": 1.353846153846154e-05,
"loss": 0.1805,
"step": 210
},
{
"epoch": 1.69,
"learning_rate": 1.3230769230769231e-05,
"loss": 0.1292,
"step": 220
},
{
"epoch": 1.77,
"learning_rate": 1.2923076923076925e-05,
"loss": 0.2089,
"step": 230
},
{
"epoch": 1.85,
"learning_rate": 1.2615384615384616e-05,
"loss": 0.1767,
"step": 240
},
{
"epoch": 1.92,
"learning_rate": 1.230769230769231e-05,
"loss": 0.2526,
"step": 250
},
{
"epoch": 2.0,
"learning_rate": 1.2e-05,
"loss": 0.2885,
"step": 260
},
{
"epoch": 2.0,
"eval_accuracy": 0.9924812030075187,
"eval_loss": 0.10981201380491257,
"eval_runtime": 1.2322,
"eval_samples_per_second": 107.933,
"eval_steps_per_second": 13.796,
"step": 260
},
{
"epoch": 2.08,
"learning_rate": 1.1692307692307694e-05,
"loss": 0.1127,
"step": 270
},
{
"epoch": 2.15,
"learning_rate": 1.1384615384615385e-05,
"loss": 0.2333,
"step": 280
},
{
"epoch": 2.23,
"learning_rate": 1.1076923076923079e-05,
"loss": 0.154,
"step": 290
},
{
"epoch": 2.31,
"learning_rate": 1.076923076923077e-05,
"loss": 0.154,
"step": 300
},
{
"epoch": 2.38,
"learning_rate": 1.0461538461538463e-05,
"loss": 0.1212,
"step": 310
},
{
"epoch": 2.46,
"learning_rate": 1.0153846153846154e-05,
"loss": 0.158,
"step": 320
},
{
"epoch": 2.54,
"learning_rate": 9.846153846153848e-06,
"loss": 0.14,
"step": 330
},
{
"epoch": 2.62,
"learning_rate": 9.53846153846154e-06,
"loss": 0.076,
"step": 340
},
{
"epoch": 2.69,
"learning_rate": 9.230769230769232e-06,
"loss": 0.1469,
"step": 350
},
{
"epoch": 2.77,
"learning_rate": 8.923076923076925e-06,
"loss": 0.079,
"step": 360
},
{
"epoch": 2.85,
"learning_rate": 8.615384615384617e-06,
"loss": 0.0893,
"step": 370
},
{
"epoch": 2.92,
"learning_rate": 8.307692307692309e-06,
"loss": 0.0895,
"step": 380
},
{
"epoch": 3.0,
"learning_rate": 8.000000000000001e-06,
"loss": 0.1555,
"step": 390
},
{
"epoch": 3.0,
"eval_accuracy": 0.9849624060150376,
"eval_loss": 0.1041199117898941,
"eval_runtime": 1.2262,
"eval_samples_per_second": 108.466,
"eval_steps_per_second": 13.864,
"step": 390
},
{
"epoch": 3.08,
"learning_rate": 7.692307692307694e-06,
"loss": 0.0724,
"step": 400
},
{
"epoch": 3.15,
"learning_rate": 7.384615384615386e-06,
"loss": 0.1879,
"step": 410
},
{
"epoch": 3.23,
"learning_rate": 7.076923076923078e-06,
"loss": 0.1089,
"step": 420
},
{
"epoch": 3.31,
"learning_rate": 6.76923076923077e-06,
"loss": 0.1225,
"step": 430
},
{
"epoch": 3.38,
"learning_rate": 6.461538461538463e-06,
"loss": 0.1831,
"step": 440
},
{
"epoch": 3.46,
"learning_rate": 6.153846153846155e-06,
"loss": 0.1386,
"step": 450
},
{
"epoch": 3.54,
"learning_rate": 5.846153846153847e-06,
"loss": 0.1456,
"step": 460
},
{
"epoch": 3.62,
"learning_rate": 5.538461538461539e-06,
"loss": 0.1152,
"step": 470
},
{
"epoch": 3.69,
"learning_rate": 5.230769230769232e-06,
"loss": 0.063,
"step": 480
},
{
"epoch": 3.77,
"learning_rate": 4.923076923076924e-06,
"loss": 0.1424,
"step": 490
},
{
"epoch": 3.85,
"learning_rate": 4.615384615384616e-06,
"loss": 0.1747,
"step": 500
},
{
"epoch": 3.92,
"learning_rate": 4.307692307692308e-06,
"loss": 0.1592,
"step": 510
},
{
"epoch": 4.0,
"learning_rate": 4.000000000000001e-06,
"loss": 0.092,
"step": 520
},
{
"epoch": 4.0,
"eval_accuracy": 0.9924812030075187,
"eval_loss": 0.08177243918180466,
"eval_runtime": 1.3043,
"eval_samples_per_second": 101.974,
"eval_steps_per_second": 13.034,
"step": 520
},
{
"epoch": 4.08,
"learning_rate": 3.692307692307693e-06,
"loss": 0.0582,
"step": 530
},
{
"epoch": 4.15,
"learning_rate": 3.384615384615385e-06,
"loss": 0.0949,
"step": 540
},
{
"epoch": 4.23,
"learning_rate": 3.0769230769230774e-06,
"loss": 0.1249,
"step": 550
},
{
"epoch": 4.31,
"learning_rate": 2.7692307692307697e-06,
"loss": 0.07,
"step": 560
},
{
"epoch": 4.38,
"learning_rate": 2.461538461538462e-06,
"loss": 0.1654,
"step": 570
},
{
"epoch": 4.46,
"learning_rate": 2.153846153846154e-06,
"loss": 0.1108,
"step": 580
},
{
"epoch": 4.54,
"learning_rate": 1.8461538461538465e-06,
"loss": 0.0476,
"step": 590
},
{
"epoch": 4.62,
"learning_rate": 1.5384615384615387e-06,
"loss": 0.117,
"step": 600
},
{
"epoch": 4.69,
"learning_rate": 1.230769230769231e-06,
"loss": 0.1404,
"step": 610
},
{
"epoch": 4.77,
"learning_rate": 9.230769230769232e-07,
"loss": 0.0927,
"step": 620
},
{
"epoch": 4.85,
"learning_rate": 6.153846153846155e-07,
"loss": 0.216,
"step": 630
},
{
"epoch": 4.92,
"learning_rate": 3.0769230769230774e-07,
"loss": 0.0965,
"step": 640
},
{
"epoch": 5.0,
"learning_rate": 0.0,
"loss": 0.1591,
"step": 650
},
{
"epoch": 5.0,
"eval_accuracy": 0.9849624060150376,
"eval_loss": 0.0826253741979599,
"eval_runtime": 1.2267,
"eval_samples_per_second": 108.418,
"eval_steps_per_second": 13.858,
"step": 650
},
{
"epoch": 5.0,
"step": 650,
"total_flos": 4.006371770595533e+17,
"train_loss": 0.0,
"train_runtime": 83.963,
"train_samples_per_second": 61.575,
"train_steps_per_second": 7.742
}
],
"max_steps": 650,
"num_train_epochs": 5,
"total_flos": 4.006371770595533e+17,
"trial_name": null,
"trial_params": null
}