FT_3 / checkpoint-1800 /trainer_state.json
sheepy928's picture
Training in progress, step 1800, checkpoint
05b0799
raw
history blame
22.7 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 9.574468085106384,
"eval_steps": 300,
"global_step": 1800,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"learning_rate": 5e-07,
"loss": 1.0874,
"step": 10
},
{
"epoch": 0.11,
"learning_rate": 1e-06,
"loss": 0.9487,
"step": 20
},
{
"epoch": 0.16,
"learning_rate": 1.5e-06,
"loss": 0.7586,
"step": 30
},
{
"epoch": 0.21,
"learning_rate": 2e-06,
"loss": 0.7225,
"step": 40
},
{
"epoch": 0.27,
"learning_rate": 2.5e-06,
"loss": 0.7364,
"step": 50
},
{
"epoch": 0.32,
"learning_rate": 3e-06,
"loss": 0.7265,
"step": 60
},
{
"epoch": 0.37,
"learning_rate": 3.5e-06,
"loss": 0.7267,
"step": 70
},
{
"epoch": 0.43,
"learning_rate": 4e-06,
"loss": 0.5697,
"step": 80
},
{
"epoch": 0.48,
"learning_rate": 4.5e-06,
"loss": 1.018,
"step": 90
},
{
"epoch": 0.53,
"learning_rate": 5e-06,
"loss": 0.7875,
"step": 100
},
{
"epoch": 0.59,
"learning_rate": 5.5e-06,
"loss": 0.8242,
"step": 110
},
{
"epoch": 0.64,
"learning_rate": 6e-06,
"loss": 0.8034,
"step": 120
},
{
"epoch": 0.69,
"learning_rate": 6.5e-06,
"loss": 0.7717,
"step": 130
},
{
"epoch": 0.74,
"learning_rate": 7e-06,
"loss": 0.8337,
"step": 140
},
{
"epoch": 0.8,
"learning_rate": 7.5e-06,
"loss": 0.6884,
"step": 150
},
{
"epoch": 0.85,
"learning_rate": 8e-06,
"loss": 0.9129,
"step": 160
},
{
"epoch": 0.9,
"learning_rate": 8.500000000000002e-06,
"loss": 0.5637,
"step": 170
},
{
"epoch": 0.96,
"learning_rate": 9e-06,
"loss": 1.0458,
"step": 180
},
{
"epoch": 1.01,
"learning_rate": 9.5e-06,
"loss": 0.9414,
"step": 190
},
{
"epoch": 1.06,
"learning_rate": 1e-05,
"loss": 0.6379,
"step": 200
},
{
"epoch": 1.12,
"learning_rate": 1.0500000000000001e-05,
"loss": 0.9249,
"step": 210
},
{
"epoch": 1.17,
"learning_rate": 1.1e-05,
"loss": 0.6944,
"step": 220
},
{
"epoch": 1.22,
"learning_rate": 1.15e-05,
"loss": 0.9221,
"step": 230
},
{
"epoch": 1.28,
"learning_rate": 1.2e-05,
"loss": 0.6475,
"step": 240
},
{
"epoch": 1.33,
"learning_rate": 1.25e-05,
"loss": 0.7748,
"step": 250
},
{
"epoch": 1.38,
"learning_rate": 1.3e-05,
"loss": 0.8705,
"step": 260
},
{
"epoch": 1.44,
"learning_rate": 1.35e-05,
"loss": 0.7737,
"step": 270
},
{
"epoch": 1.49,
"learning_rate": 1.4e-05,
"loss": 0.8643,
"step": 280
},
{
"epoch": 1.54,
"learning_rate": 1.4500000000000002e-05,
"loss": 0.8428,
"step": 290
},
{
"epoch": 1.6,
"learning_rate": 1.5e-05,
"loss": 0.6785,
"step": 300
},
{
"epoch": 1.6,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.7930460572242737,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 6.0663,
"eval_samples_per_second": 247.266,
"eval_steps_per_second": 7.748,
"step": 300
},
{
"epoch": 1.65,
"learning_rate": 1.55e-05,
"loss": 0.6076,
"step": 310
},
{
"epoch": 1.7,
"learning_rate": 1.6e-05,
"loss": 0.5963,
"step": 320
},
{
"epoch": 1.76,
"learning_rate": 1.65e-05,
"loss": 0.6626,
"step": 330
},
{
"epoch": 1.81,
"learning_rate": 1.7000000000000003e-05,
"loss": 0.8379,
"step": 340
},
{
"epoch": 1.86,
"learning_rate": 1.7500000000000002e-05,
"loss": 0.8851,
"step": 350
},
{
"epoch": 1.91,
"learning_rate": 1.8e-05,
"loss": 0.7489,
"step": 360
},
{
"epoch": 1.97,
"learning_rate": 1.85e-05,
"loss": 0.7573,
"step": 370
},
{
"epoch": 2.02,
"learning_rate": 1.9e-05,
"loss": 0.8018,
"step": 380
},
{
"epoch": 2.07,
"learning_rate": 1.95e-05,
"loss": 0.6645,
"step": 390
},
{
"epoch": 2.13,
"learning_rate": 2e-05,
"loss": 0.8677,
"step": 400
},
{
"epoch": 2.18,
"learning_rate": 2.05e-05,
"loss": 0.7478,
"step": 410
},
{
"epoch": 2.23,
"learning_rate": 2.1000000000000002e-05,
"loss": 0.8551,
"step": 420
},
{
"epoch": 2.29,
"learning_rate": 2.1499999999999997e-05,
"loss": 0.9323,
"step": 430
},
{
"epoch": 2.34,
"learning_rate": 2.2e-05,
"loss": 0.7536,
"step": 440
},
{
"epoch": 2.39,
"learning_rate": 2.2499999999999998e-05,
"loss": 0.5336,
"step": 450
},
{
"epoch": 2.45,
"learning_rate": 2.3e-05,
"loss": 0.8955,
"step": 460
},
{
"epoch": 2.5,
"learning_rate": 2.3500000000000002e-05,
"loss": 0.7926,
"step": 470
},
{
"epoch": 2.55,
"learning_rate": 2.4e-05,
"loss": 0.5713,
"step": 480
},
{
"epoch": 2.61,
"learning_rate": 2.4500000000000003e-05,
"loss": 0.8568,
"step": 490
},
{
"epoch": 2.66,
"learning_rate": 2.5e-05,
"loss": 0.6348,
"step": 500
},
{
"epoch": 2.71,
"learning_rate": 2.55e-05,
"loss": 0.6223,
"step": 510
},
{
"epoch": 2.77,
"learning_rate": 2.6e-05,
"loss": 0.7579,
"step": 520
},
{
"epoch": 2.82,
"learning_rate": 2.65e-05,
"loss": 0.6325,
"step": 530
},
{
"epoch": 2.87,
"learning_rate": 2.7e-05,
"loss": 0.7276,
"step": 540
},
{
"epoch": 2.93,
"learning_rate": 2.75e-05,
"loss": 0.8766,
"step": 550
},
{
"epoch": 2.98,
"learning_rate": 2.8e-05,
"loss": 0.7107,
"step": 560
},
{
"epoch": 3.03,
"learning_rate": 2.85e-05,
"loss": 0.5904,
"step": 570
},
{
"epoch": 3.09,
"learning_rate": 2.9000000000000004e-05,
"loss": 0.7125,
"step": 580
},
{
"epoch": 3.14,
"learning_rate": 2.95e-05,
"loss": 0.755,
"step": 590
},
{
"epoch": 3.19,
"learning_rate": 3e-05,
"loss": 0.5583,
"step": 600
},
{
"epoch": 3.19,
"eval_accuracy": 0.7613333333333333,
"eval_combined_score": 0.7403762607674487,
"eval_f1": 0.7316181693637903,
"eval_loss": 0.6910097599029541,
"eval_precision": 0.7072202070393374,
"eval_recall": 0.7613333333333333,
"eval_runtime": 6.0882,
"eval_samples_per_second": 246.376,
"eval_steps_per_second": 7.72,
"step": 600
},
{
"epoch": 3.24,
"learning_rate": 3.05e-05,
"loss": 0.6527,
"step": 610
},
{
"epoch": 3.3,
"learning_rate": 3.1e-05,
"loss": 0.7676,
"step": 620
},
{
"epoch": 3.35,
"learning_rate": 3.15e-05,
"loss": 0.766,
"step": 630
},
{
"epoch": 3.4,
"learning_rate": 3.2e-05,
"loss": 0.6217,
"step": 640
},
{
"epoch": 3.46,
"learning_rate": 3.2500000000000004e-05,
"loss": 0.5405,
"step": 650
},
{
"epoch": 3.51,
"learning_rate": 3.3e-05,
"loss": 0.7951,
"step": 660
},
{
"epoch": 3.56,
"learning_rate": 3.35e-05,
"loss": 0.552,
"step": 670
},
{
"epoch": 3.62,
"learning_rate": 3.4000000000000007e-05,
"loss": 0.6521,
"step": 680
},
{
"epoch": 3.67,
"learning_rate": 3.4500000000000005e-05,
"loss": 0.6814,
"step": 690
},
{
"epoch": 3.72,
"learning_rate": 3.5000000000000004e-05,
"loss": 0.3992,
"step": 700
},
{
"epoch": 3.78,
"learning_rate": 3.5499999999999996e-05,
"loss": 0.7018,
"step": 710
},
{
"epoch": 3.83,
"learning_rate": 3.6e-05,
"loss": 0.664,
"step": 720
},
{
"epoch": 3.88,
"learning_rate": 3.65e-05,
"loss": 0.6891,
"step": 730
},
{
"epoch": 3.94,
"learning_rate": 3.7e-05,
"loss": 0.6063,
"step": 740
},
{
"epoch": 3.99,
"learning_rate": 3.75e-05,
"loss": 0.6543,
"step": 750
},
{
"epoch": 4.04,
"learning_rate": 3.8e-05,
"loss": 0.5638,
"step": 760
},
{
"epoch": 4.1,
"learning_rate": 3.85e-05,
"loss": 0.6984,
"step": 770
},
{
"epoch": 4.15,
"learning_rate": 3.9e-05,
"loss": 0.7148,
"step": 780
},
{
"epoch": 4.2,
"learning_rate": 3.95e-05,
"loss": 0.4915,
"step": 790
},
{
"epoch": 4.26,
"learning_rate": 4e-05,
"loss": 0.648,
"step": 800
},
{
"epoch": 4.31,
"learning_rate": 4.05e-05,
"loss": 0.4527,
"step": 810
},
{
"epoch": 4.36,
"learning_rate": 4.1e-05,
"loss": 0.551,
"step": 820
},
{
"epoch": 4.41,
"learning_rate": 4.1500000000000006e-05,
"loss": 0.5909,
"step": 830
},
{
"epoch": 4.47,
"learning_rate": 4.2000000000000004e-05,
"loss": 0.603,
"step": 840
},
{
"epoch": 4.52,
"learning_rate": 4.25e-05,
"loss": 1.0745,
"step": 850
},
{
"epoch": 4.57,
"learning_rate": 4.2999999999999995e-05,
"loss": 0.6697,
"step": 860
},
{
"epoch": 4.63,
"learning_rate": 4.35e-05,
"loss": 0.8374,
"step": 870
},
{
"epoch": 4.68,
"learning_rate": 4.4e-05,
"loss": 0.4891,
"step": 880
},
{
"epoch": 4.73,
"learning_rate": 4.45e-05,
"loss": 0.5956,
"step": 890
},
{
"epoch": 4.79,
"learning_rate": 4.4999999999999996e-05,
"loss": 0.7857,
"step": 900
},
{
"epoch": 4.79,
"eval_accuracy": 0.7386666666666667,
"eval_combined_score": 0.6626504648943422,
"eval_f1": 0.6276400817995911,
"eval_loss": 0.6514685153961182,
"eval_precision": 0.5456284444444445,
"eval_recall": 0.7386666666666667,
"eval_runtime": 5.9473,
"eval_samples_per_second": 252.216,
"eval_steps_per_second": 7.903,
"step": 900
},
{
"epoch": 4.84,
"learning_rate": 4.55e-05,
"loss": 0.8566,
"step": 910
},
{
"epoch": 4.89,
"learning_rate": 4.6e-05,
"loss": 0.4698,
"step": 920
},
{
"epoch": 4.95,
"learning_rate": 4.65e-05,
"loss": 0.7224,
"step": 930
},
{
"epoch": 5.0,
"learning_rate": 4.7000000000000004e-05,
"loss": 0.5879,
"step": 940
},
{
"epoch": 5.05,
"learning_rate": 4.75e-05,
"loss": 0.3592,
"step": 950
},
{
"epoch": 5.11,
"learning_rate": 4.8e-05,
"loss": 0.513,
"step": 960
},
{
"epoch": 5.16,
"learning_rate": 4.85e-05,
"loss": 0.6167,
"step": 970
},
{
"epoch": 5.21,
"learning_rate": 4.9000000000000005e-05,
"loss": 0.5706,
"step": 980
},
{
"epoch": 5.27,
"learning_rate": 4.9500000000000004e-05,
"loss": 0.5523,
"step": 990
},
{
"epoch": 5.32,
"learning_rate": 5e-05,
"loss": 0.6565,
"step": 1000
},
{
"epoch": 5.37,
"learning_rate": 5.05e-05,
"loss": 0.4634,
"step": 1010
},
{
"epoch": 5.43,
"learning_rate": 5.1e-05,
"loss": 0.634,
"step": 1020
},
{
"epoch": 5.48,
"learning_rate": 5.15e-05,
"loss": 0.5472,
"step": 1030
},
{
"epoch": 5.53,
"learning_rate": 5.2e-05,
"loss": 0.595,
"step": 1040
},
{
"epoch": 5.59,
"learning_rate": 5.25e-05,
"loss": 0.4889,
"step": 1050
},
{
"epoch": 5.64,
"learning_rate": 5.3e-05,
"loss": 0.5884,
"step": 1060
},
{
"epoch": 5.69,
"learning_rate": 5.35e-05,
"loss": 0.4149,
"step": 1070
},
{
"epoch": 5.74,
"learning_rate": 5.4e-05,
"loss": 0.7573,
"step": 1080
},
{
"epoch": 5.8,
"learning_rate": 5.45e-05,
"loss": 0.599,
"step": 1090
},
{
"epoch": 5.85,
"learning_rate": 5.5e-05,
"loss": 0.366,
"step": 1100
},
{
"epoch": 5.9,
"learning_rate": 5.55e-05,
"loss": 0.4906,
"step": 1110
},
{
"epoch": 5.96,
"learning_rate": 5.6e-05,
"loss": 0.4378,
"step": 1120
},
{
"epoch": 6.01,
"learning_rate": 5.6500000000000005e-05,
"loss": 0.4886,
"step": 1130
},
{
"epoch": 6.06,
"learning_rate": 5.7e-05,
"loss": 0.342,
"step": 1140
},
{
"epoch": 6.12,
"learning_rate": 5.75e-05,
"loss": 0.7803,
"step": 1150
},
{
"epoch": 6.17,
"learning_rate": 5.800000000000001e-05,
"loss": 0.3435,
"step": 1160
},
{
"epoch": 6.22,
"learning_rate": 5.8500000000000006e-05,
"loss": 0.6149,
"step": 1170
},
{
"epoch": 6.28,
"learning_rate": 5.9e-05,
"loss": 0.3026,
"step": 1180
},
{
"epoch": 6.33,
"learning_rate": 5.9499999999999996e-05,
"loss": 0.9387,
"step": 1190
},
{
"epoch": 6.38,
"learning_rate": 6e-05,
"loss": 0.6309,
"step": 1200
},
{
"epoch": 6.38,
"eval_accuracy": 0.848,
"eval_combined_score": 0.8403047916631027,
"eval_f1": 0.8270449473875959,
"eval_loss": 0.5592399835586548,
"eval_precision": 0.8381742192648148,
"eval_recall": 0.848,
"eval_runtime": 5.9701,
"eval_samples_per_second": 251.251,
"eval_steps_per_second": 7.873,
"step": 1200
},
{
"epoch": 6.44,
"learning_rate": 6.05e-05,
"loss": 0.4672,
"step": 1210
},
{
"epoch": 6.49,
"learning_rate": 6.1e-05,
"loss": 0.4909,
"step": 1220
},
{
"epoch": 6.54,
"learning_rate": 6.15e-05,
"loss": 0.5007,
"step": 1230
},
{
"epoch": 6.6,
"learning_rate": 6.2e-05,
"loss": 0.4318,
"step": 1240
},
{
"epoch": 6.65,
"learning_rate": 6.25e-05,
"loss": 0.3623,
"step": 1250
},
{
"epoch": 6.7,
"learning_rate": 6.3e-05,
"loss": 0.3294,
"step": 1260
},
{
"epoch": 6.76,
"learning_rate": 6.35e-05,
"loss": 0.7501,
"step": 1270
},
{
"epoch": 6.81,
"learning_rate": 6.4e-05,
"loss": 0.6436,
"step": 1280
},
{
"epoch": 6.86,
"learning_rate": 6.450000000000001e-05,
"loss": 0.525,
"step": 1290
},
{
"epoch": 6.91,
"learning_rate": 6.500000000000001e-05,
"loss": 0.6047,
"step": 1300
},
{
"epoch": 6.97,
"learning_rate": 6.55e-05,
"loss": 0.3636,
"step": 1310
},
{
"epoch": 7.02,
"learning_rate": 6.6e-05,
"loss": 0.5634,
"step": 1320
},
{
"epoch": 7.07,
"learning_rate": 6.65e-05,
"loss": 0.3846,
"step": 1330
},
{
"epoch": 7.13,
"learning_rate": 6.7e-05,
"loss": 0.8436,
"step": 1340
},
{
"epoch": 7.18,
"learning_rate": 6.75e-05,
"loss": 0.4762,
"step": 1350
},
{
"epoch": 7.23,
"learning_rate": 6.800000000000001e-05,
"loss": 0.5856,
"step": 1360
},
{
"epoch": 7.29,
"learning_rate": 6.850000000000001e-05,
"loss": 0.4042,
"step": 1370
},
{
"epoch": 7.34,
"learning_rate": 6.900000000000001e-05,
"loss": 0.475,
"step": 1380
},
{
"epoch": 7.39,
"learning_rate": 6.950000000000001e-05,
"loss": 0.3535,
"step": 1390
},
{
"epoch": 7.45,
"learning_rate": 7.000000000000001e-05,
"loss": 0.2399,
"step": 1400
},
{
"epoch": 7.5,
"learning_rate": 7.049999999999999e-05,
"loss": 0.316,
"step": 1410
},
{
"epoch": 7.55,
"learning_rate": 7.095e-05,
"loss": 0.438,
"step": 1420
},
{
"epoch": 7.61,
"learning_rate": 7.145e-05,
"loss": 0.2673,
"step": 1430
},
{
"epoch": 7.66,
"learning_rate": 7.195e-05,
"loss": 0.4286,
"step": 1440
},
{
"epoch": 7.71,
"learning_rate": 7.245e-05,
"loss": 0.3655,
"step": 1450
},
{
"epoch": 7.77,
"learning_rate": 7.295e-05,
"loss": 0.434,
"step": 1460
},
{
"epoch": 7.82,
"learning_rate": 7.345e-05,
"loss": 0.6143,
"step": 1470
},
{
"epoch": 7.87,
"learning_rate": 7.395000000000001e-05,
"loss": 0.5471,
"step": 1480
},
{
"epoch": 7.93,
"learning_rate": 7.445000000000001e-05,
"loss": 0.299,
"step": 1490
},
{
"epoch": 7.98,
"learning_rate": 7.495e-05,
"loss": 0.2216,
"step": 1500
},
{
"epoch": 7.98,
"eval_accuracy": 0.8773333333333333,
"eval_combined_score": 0.8618647537688802,
"eval_f1": 0.8431508113173474,
"eval_loss": 0.5708244442939758,
"eval_precision": 0.8496415370915067,
"eval_recall": 0.8773333333333333,
"eval_runtime": 6.0286,
"eval_samples_per_second": 248.816,
"eval_steps_per_second": 7.796,
"step": 1500
},
{
"epoch": 8.03,
"learning_rate": 7.545e-05,
"loss": 0.5517,
"step": 1510
},
{
"epoch": 8.09,
"learning_rate": 7.595e-05,
"loss": 0.3237,
"step": 1520
},
{
"epoch": 8.14,
"learning_rate": 7.645e-05,
"loss": 0.3945,
"step": 1530
},
{
"epoch": 8.19,
"learning_rate": 7.695e-05,
"loss": 0.3996,
"step": 1540
},
{
"epoch": 8.24,
"learning_rate": 7.745000000000001e-05,
"loss": 0.3644,
"step": 1550
},
{
"epoch": 8.3,
"learning_rate": 7.795000000000001e-05,
"loss": 0.6767,
"step": 1560
},
{
"epoch": 8.35,
"learning_rate": 7.845000000000001e-05,
"loss": 0.434,
"step": 1570
},
{
"epoch": 8.4,
"learning_rate": 7.895000000000001e-05,
"loss": 0.4047,
"step": 1580
},
{
"epoch": 8.46,
"learning_rate": 7.945e-05,
"loss": 0.3844,
"step": 1590
},
{
"epoch": 8.51,
"learning_rate": 7.994999999999999e-05,
"loss": 0.2656,
"step": 1600
},
{
"epoch": 8.56,
"learning_rate": 8.044999999999999e-05,
"loss": 0.3318,
"step": 1610
},
{
"epoch": 8.62,
"learning_rate": 8.095e-05,
"loss": 0.4627,
"step": 1620
},
{
"epoch": 8.67,
"learning_rate": 8.145e-05,
"loss": 0.6657,
"step": 1630
},
{
"epoch": 8.72,
"learning_rate": 8.195e-05,
"loss": 0.353,
"step": 1640
},
{
"epoch": 8.78,
"learning_rate": 8.245e-05,
"loss": 0.4543,
"step": 1650
},
{
"epoch": 8.83,
"learning_rate": 8.295e-05,
"loss": 0.4121,
"step": 1660
},
{
"epoch": 8.88,
"learning_rate": 8.340000000000001e-05,
"loss": 0.2792,
"step": 1670
},
{
"epoch": 8.94,
"learning_rate": 8.39e-05,
"loss": 0.3618,
"step": 1680
},
{
"epoch": 8.99,
"learning_rate": 8.44e-05,
"loss": 0.4922,
"step": 1690
},
{
"epoch": 9.04,
"learning_rate": 8.49e-05,
"loss": 0.6822,
"step": 1700
},
{
"epoch": 9.1,
"learning_rate": 8.54e-05,
"loss": 0.4936,
"step": 1710
},
{
"epoch": 9.15,
"learning_rate": 8.59e-05,
"loss": 0.4652,
"step": 1720
},
{
"epoch": 9.2,
"learning_rate": 8.640000000000001e-05,
"loss": 0.3529,
"step": 1730
},
{
"epoch": 9.26,
"learning_rate": 8.690000000000001e-05,
"loss": 0.5115,
"step": 1740
},
{
"epoch": 9.31,
"learning_rate": 8.740000000000001e-05,
"loss": 0.5859,
"step": 1750
},
{
"epoch": 9.36,
"learning_rate": 8.790000000000001e-05,
"loss": 0.5069,
"step": 1760
},
{
"epoch": 9.41,
"learning_rate": 8.840000000000001e-05,
"loss": 0.2584,
"step": 1770
},
{
"epoch": 9.47,
"learning_rate": 8.885e-05,
"loss": 0.313,
"step": 1780
},
{
"epoch": 9.52,
"learning_rate": 8.935e-05,
"loss": 0.4103,
"step": 1790
},
{
"epoch": 9.57,
"learning_rate": 8.985e-05,
"loss": 0.3214,
"step": 1800
},
{
"epoch": 9.57,
"eval_accuracy": 0.896,
"eval_combined_score": 0.8819938675326909,
"eval_f1": 0.8583787127091727,
"eval_loss": 0.45495954155921936,
"eval_precision": 0.8775967574215913,
"eval_recall": 0.896,
"eval_runtime": 6.1706,
"eval_samples_per_second": 243.087,
"eval_steps_per_second": 7.617,
"step": 1800
}
],
"logging_steps": 10,
"max_steps": 5640,
"num_train_epochs": 30,
"save_steps": 300,
"total_flos": 3779360999866368.0,
"trial_name": null,
"trial_params": null
}