SLM_vs_LLM_experiments
/
max_seq_length_128_experiments
/LoRA
/Qwen
/Qwen1.5_7B_LoRA_coastalcph
/lex_glue_ledgar
/checkpoint-3700
/trainer_state.json
{ | |
"best_metric": 0.5025032162666321, | |
"best_model_checkpoint": "../experiments_checkpoints/LoRA/Qwen/Qwen1.5_7B_LoRA_coastalcph/lex_glue_ledgar/checkpoint-3700", | |
"epoch": 1.9733333333333334, | |
"eval_steps": 100, | |
"global_step": 3700, | |
"is_hyper_param_search": false, | |
"is_local_process_zero": true, | |
"is_world_process_zero": true, | |
"log_history": [ | |
{ | |
"epoch": 0.01, | |
"grad_norm": 82.343017578125, | |
"learning_rate": 4.991111111111111e-05, | |
"loss": 11.1109, | |
"step": 10 | |
}, | |
{ | |
"epoch": 0.01, | |
"grad_norm": 83.22059631347656, | |
"learning_rate": 4.982222222222222e-05, | |
"loss": 8.525, | |
"step": 20 | |
}, | |
{ | |
"epoch": 0.02, | |
"grad_norm": 81.96794891357422, | |
"learning_rate": 4.973333333333334e-05, | |
"loss": 6.6453, | |
"step": 30 | |
}, | |
{ | |
"epoch": 0.02, | |
"grad_norm": 71.0719985961914, | |
"learning_rate": 4.964444444444445e-05, | |
"loss": 5.1352, | |
"step": 40 | |
}, | |
{ | |
"epoch": 0.03, | |
"grad_norm": 69.25579071044922, | |
"learning_rate": 4.955555555555556e-05, | |
"loss": 3.8117, | |
"step": 50 | |
}, | |
{ | |
"epoch": 0.03, | |
"grad_norm": 62.54920959472656, | |
"learning_rate": 4.9466666666666665e-05, | |
"loss": 2.7141, | |
"step": 60 | |
}, | |
{ | |
"epoch": 0.04, | |
"grad_norm": 51.00497055053711, | |
"learning_rate": 4.9377777777777776e-05, | |
"loss": 2.4369, | |
"step": 70 | |
}, | |
{ | |
"epoch": 0.04, | |
"grad_norm": 66.0743637084961, | |
"learning_rate": 4.928888888888889e-05, | |
"loss": 1.966, | |
"step": 80 | |
}, | |
{ | |
"epoch": 0.05, | |
"grad_norm": 55.161354064941406, | |
"learning_rate": 4.92e-05, | |
"loss": 1.8213, | |
"step": 90 | |
}, | |
{ | |
"epoch": 0.05, | |
"grad_norm": 50.841796875, | |
"learning_rate": 4.9111111111111114e-05, | |
"loss": 1.7995, | |
"step": 100 | |
}, | |
{ | |
"epoch": 0.05, | |
"eval_accuracy": 0.6512, | |
"eval_f1_macro": 0.4676003614917378, | |
"eval_f1_micro": 0.6512, | |
"eval_loss": 1.6894375085830688, | |
"eval_runtime": 126.3135, | |
"eval_samples_per_second": 79.168, | |
"eval_steps_per_second": 2.478, | |
"step": 100 | |
}, | |
{ | |
"epoch": 0.06, | |
"grad_norm": 44.830013275146484, | |
"learning_rate": 4.9022222222222224e-05, | |
"loss": 1.7258, | |
"step": 110 | |
}, | |
{ | |
"epoch": 0.06, | |
"grad_norm": 37.19465637207031, | |
"learning_rate": 4.8933333333333335e-05, | |
"loss": 1.4613, | |
"step": 120 | |
}, | |
{ | |
"epoch": 0.07, | |
"grad_norm": 50.72254943847656, | |
"learning_rate": 4.8844444444444445e-05, | |
"loss": 1.4131, | |
"step": 130 | |
}, | |
{ | |
"epoch": 0.07, | |
"grad_norm": 49.2353630065918, | |
"learning_rate": 4.875555555555556e-05, | |
"loss": 1.4369, | |
"step": 140 | |
}, | |
{ | |
"epoch": 0.08, | |
"grad_norm": 44.442604064941406, | |
"learning_rate": 4.866666666666667e-05, | |
"loss": 1.2461, | |
"step": 150 | |
}, | |
{ | |
"epoch": 0.09, | |
"grad_norm": 54.64115524291992, | |
"learning_rate": 4.8577777777777776e-05, | |
"loss": 1.2523, | |
"step": 160 | |
}, | |
{ | |
"epoch": 0.09, | |
"grad_norm": 46.350425720214844, | |
"learning_rate": 4.848888888888889e-05, | |
"loss": 1.2385, | |
"step": 170 | |
}, | |
{ | |
"epoch": 0.1, | |
"grad_norm": 45.80131912231445, | |
"learning_rate": 4.8400000000000004e-05, | |
"loss": 1.2934, | |
"step": 180 | |
}, | |
{ | |
"epoch": 0.1, | |
"grad_norm": 43.66839599609375, | |
"learning_rate": 4.8311111111111115e-05, | |
"loss": 1.1484, | |
"step": 190 | |
}, | |
{ | |
"epoch": 0.11, | |
"grad_norm": 45.98219680786133, | |
"learning_rate": 4.8222222222222225e-05, | |
"loss": 1.3922, | |
"step": 200 | |
}, | |
{ | |
"epoch": 0.11, | |
"eval_accuracy": 0.7076, | |
"eval_f1_macro": 0.5868340973418246, | |
"eval_f1_micro": 0.7076, | |
"eval_loss": 1.2208046913146973, | |
"eval_runtime": 125.6037, | |
"eval_samples_per_second": 79.616, | |
"eval_steps_per_second": 2.492, | |
"step": 200 | |
}, | |
{ | |
"epoch": 0.11, | |
"grad_norm": 53.35915756225586, | |
"learning_rate": 4.8133333333333336e-05, | |
"loss": 1.2534, | |
"step": 210 | |
}, | |
{ | |
"epoch": 0.12, | |
"grad_norm": 49.47136688232422, | |
"learning_rate": 4.8044444444444446e-05, | |
"loss": 1.2297, | |
"step": 220 | |
}, | |
{ | |
"epoch": 0.12, | |
"grad_norm": 41.21822738647461, | |
"learning_rate": 4.7955555555555556e-05, | |
"loss": 1.136, | |
"step": 230 | |
}, | |
{ | |
"epoch": 0.13, | |
"grad_norm": 49.57373809814453, | |
"learning_rate": 4.7866666666666674e-05, | |
"loss": 1.1995, | |
"step": 240 | |
}, | |
{ | |
"epoch": 0.13, | |
"grad_norm": 60.3203010559082, | |
"learning_rate": 4.7777777777777784e-05, | |
"loss": 1.1587, | |
"step": 250 | |
}, | |
{ | |
"epoch": 0.14, | |
"grad_norm": 44.72378158569336, | |
"learning_rate": 4.768888888888889e-05, | |
"loss": 1.1577, | |
"step": 260 | |
}, | |
{ | |
"epoch": 0.14, | |
"grad_norm": 45.23435974121094, | |
"learning_rate": 4.76e-05, | |
"loss": 1.0536, | |
"step": 270 | |
}, | |
{ | |
"epoch": 0.15, | |
"grad_norm": 34.0523567199707, | |
"learning_rate": 4.751111111111111e-05, | |
"loss": 1.0193, | |
"step": 280 | |
}, | |
{ | |
"epoch": 0.15, | |
"grad_norm": 43.04400634765625, | |
"learning_rate": 4.7422222222222226e-05, | |
"loss": 1.0108, | |
"step": 290 | |
}, | |
{ | |
"epoch": 0.16, | |
"grad_norm": 45.1166877746582, | |
"learning_rate": 4.7333333333333336e-05, | |
"loss": 1.0552, | |
"step": 300 | |
}, | |
{ | |
"epoch": 0.16, | |
"eval_accuracy": 0.7634, | |
"eval_f1_macro": 0.6328915988513745, | |
"eval_f1_micro": 0.7634, | |
"eval_loss": 0.9664570093154907, | |
"eval_runtime": 125.4379, | |
"eval_samples_per_second": 79.721, | |
"eval_steps_per_second": 2.495, | |
"step": 300 | |
}, | |
{ | |
"epoch": 0.17, | |
"grad_norm": 42.98120880126953, | |
"learning_rate": 4.724444444444445e-05, | |
"loss": 0.9302, | |
"step": 310 | |
}, | |
{ | |
"epoch": 0.17, | |
"grad_norm": 32.83863830566406, | |
"learning_rate": 4.715555555555556e-05, | |
"loss": 0.8612, | |
"step": 320 | |
}, | |
{ | |
"epoch": 0.18, | |
"grad_norm": 51.440391540527344, | |
"learning_rate": 4.706666666666667e-05, | |
"loss": 1.2096, | |
"step": 330 | |
}, | |
{ | |
"epoch": 0.18, | |
"grad_norm": 32.38331604003906, | |
"learning_rate": 4.6977777777777785e-05, | |
"loss": 0.939, | |
"step": 340 | |
}, | |
{ | |
"epoch": 0.19, | |
"grad_norm": 32.610382080078125, | |
"learning_rate": 4.6888888888888895e-05, | |
"loss": 0.9571, | |
"step": 350 | |
}, | |
{ | |
"epoch": 0.19, | |
"grad_norm": 49.74464797973633, | |
"learning_rate": 4.6800000000000006e-05, | |
"loss": 1.0897, | |
"step": 360 | |
}, | |
{ | |
"epoch": 0.2, | |
"grad_norm": 40.764678955078125, | |
"learning_rate": 4.671111111111111e-05, | |
"loss": 0.85, | |
"step": 370 | |
}, | |
{ | |
"epoch": 0.2, | |
"grad_norm": 55.797000885009766, | |
"learning_rate": 4.662222222222222e-05, | |
"loss": 0.9105, | |
"step": 380 | |
}, | |
{ | |
"epoch": 0.21, | |
"grad_norm": 41.21658706665039, | |
"learning_rate": 4.653333333333334e-05, | |
"loss": 0.8054, | |
"step": 390 | |
}, | |
{ | |
"epoch": 0.21, | |
"grad_norm": 39.946632385253906, | |
"learning_rate": 4.644444444444445e-05, | |
"loss": 0.8416, | |
"step": 400 | |
}, | |
{ | |
"epoch": 0.21, | |
"eval_accuracy": 0.767, | |
"eval_f1_macro": 0.6280418494347138, | |
"eval_f1_micro": 0.767, | |
"eval_loss": 0.9614554643630981, | |
"eval_runtime": 125.3428, | |
"eval_samples_per_second": 79.781, | |
"eval_steps_per_second": 2.497, | |
"step": 400 | |
}, | |
{ | |
"epoch": 0.22, | |
"grad_norm": 40.86934280395508, | |
"learning_rate": 4.635555555555556e-05, | |
"loss": 1.0155, | |
"step": 410 | |
}, | |
{ | |
"epoch": 0.22, | |
"grad_norm": 39.81444549560547, | |
"learning_rate": 4.626666666666667e-05, | |
"loss": 0.9998, | |
"step": 420 | |
}, | |
{ | |
"epoch": 0.23, | |
"grad_norm": 40.45600509643555, | |
"learning_rate": 4.617777777777778e-05, | |
"loss": 0.7995, | |
"step": 430 | |
}, | |
{ | |
"epoch": 0.23, | |
"grad_norm": 47.23322677612305, | |
"learning_rate": 4.608888888888889e-05, | |
"loss": 0.6886, | |
"step": 440 | |
}, | |
{ | |
"epoch": 0.24, | |
"grad_norm": 45.51860046386719, | |
"learning_rate": 4.600000000000001e-05, | |
"loss": 0.7401, | |
"step": 450 | |
}, | |
{ | |
"epoch": 0.25, | |
"grad_norm": 31.879140853881836, | |
"learning_rate": 4.591111111111112e-05, | |
"loss": 0.9815, | |
"step": 460 | |
}, | |
{ | |
"epoch": 0.25, | |
"grad_norm": 38.960289001464844, | |
"learning_rate": 4.582222222222222e-05, | |
"loss": 1.0263, | |
"step": 470 | |
}, | |
{ | |
"epoch": 0.26, | |
"grad_norm": 26.54925537109375, | |
"learning_rate": 4.573333333333333e-05, | |
"loss": 0.8351, | |
"step": 480 | |
}, | |
{ | |
"epoch": 0.26, | |
"grad_norm": 42.07994842529297, | |
"learning_rate": 4.564444444444444e-05, | |
"loss": 0.8951, | |
"step": 490 | |
}, | |
{ | |
"epoch": 0.27, | |
"grad_norm": 43.244110107421875, | |
"learning_rate": 4.555555555555556e-05, | |
"loss": 0.8204, | |
"step": 500 | |
}, | |
{ | |
"epoch": 0.27, | |
"eval_accuracy": 0.7892, | |
"eval_f1_macro": 0.6680353341510004, | |
"eval_f1_micro": 0.7892, | |
"eval_loss": 0.846860945224762, | |
"eval_runtime": 125.1499, | |
"eval_samples_per_second": 79.904, | |
"eval_steps_per_second": 2.501, | |
"step": 500 | |
}, | |
{ | |
"epoch": 0.27, | |
"grad_norm": 40.45793151855469, | |
"learning_rate": 4.546666666666667e-05, | |
"loss": 0.8416, | |
"step": 510 | |
}, | |
{ | |
"epoch": 0.28, | |
"grad_norm": 48.35597610473633, | |
"learning_rate": 4.537777777777778e-05, | |
"loss": 0.8184, | |
"step": 520 | |
}, | |
{ | |
"epoch": 0.28, | |
"grad_norm": 38.001399993896484, | |
"learning_rate": 4.528888888888889e-05, | |
"loss": 0.8784, | |
"step": 530 | |
}, | |
{ | |
"epoch": 0.29, | |
"grad_norm": 35.90673065185547, | |
"learning_rate": 4.52e-05, | |
"loss": 0.8764, | |
"step": 540 | |
}, | |
{ | |
"epoch": 0.29, | |
"grad_norm": 42.99267578125, | |
"learning_rate": 4.511111111111112e-05, | |
"loss": 0.7643, | |
"step": 550 | |
}, | |
{ | |
"epoch": 0.3, | |
"grad_norm": 35.354740142822266, | |
"learning_rate": 4.502222222222223e-05, | |
"loss": 0.7455, | |
"step": 560 | |
}, | |
{ | |
"epoch": 0.3, | |
"grad_norm": 46.95038986206055, | |
"learning_rate": 4.493333333333333e-05, | |
"loss": 1.0796, | |
"step": 570 | |
}, | |
{ | |
"epoch": 0.31, | |
"grad_norm": 36.5641975402832, | |
"learning_rate": 4.484444444444444e-05, | |
"loss": 0.7303, | |
"step": 580 | |
}, | |
{ | |
"epoch": 0.31, | |
"grad_norm": 31.755895614624023, | |
"learning_rate": 4.475555555555555e-05, | |
"loss": 0.6932, | |
"step": 590 | |
}, | |
{ | |
"epoch": 0.32, | |
"grad_norm": 41.90700149536133, | |
"learning_rate": 4.466666666666667e-05, | |
"loss": 0.7359, | |
"step": 600 | |
}, | |
{ | |
"epoch": 0.32, | |
"eval_accuracy": 0.8025, | |
"eval_f1_macro": 0.6858714672474995, | |
"eval_f1_micro": 0.8025, | |
"eval_loss": 0.7820109128952026, | |
"eval_runtime": 125.0637, | |
"eval_samples_per_second": 79.959, | |
"eval_steps_per_second": 2.503, | |
"step": 600 | |
}, | |
{ | |
"epoch": 0.33, | |
"grad_norm": 40.491641998291016, | |
"learning_rate": 4.457777777777778e-05, | |
"loss": 0.7398, | |
"step": 610 | |
}, | |
{ | |
"epoch": 0.33, | |
"grad_norm": 40.14454650878906, | |
"learning_rate": 4.448888888888889e-05, | |
"loss": 0.7974, | |
"step": 620 | |
}, | |
{ | |
"epoch": 0.34, | |
"grad_norm": 39.546302795410156, | |
"learning_rate": 4.44e-05, | |
"loss": 0.6925, | |
"step": 630 | |
}, | |
{ | |
"epoch": 0.34, | |
"grad_norm": 27.2608699798584, | |
"learning_rate": 4.431111111111111e-05, | |
"loss": 0.8556, | |
"step": 640 | |
}, | |
{ | |
"epoch": 0.35, | |
"grad_norm": 28.21966552734375, | |
"learning_rate": 4.422222222222222e-05, | |
"loss": 0.798, | |
"step": 650 | |
}, | |
{ | |
"epoch": 0.35, | |
"grad_norm": 26.93427276611328, | |
"learning_rate": 4.413333333333334e-05, | |
"loss": 0.8146, | |
"step": 660 | |
}, | |
{ | |
"epoch": 0.36, | |
"grad_norm": 38.73881912231445, | |
"learning_rate": 4.404444444444445e-05, | |
"loss": 0.7848, | |
"step": 670 | |
}, | |
{ | |
"epoch": 0.36, | |
"grad_norm": 29.951839447021484, | |
"learning_rate": 4.3955555555555554e-05, | |
"loss": 0.6993, | |
"step": 680 | |
}, | |
{ | |
"epoch": 0.37, | |
"grad_norm": 39.4987678527832, | |
"learning_rate": 4.3866666666666665e-05, | |
"loss": 0.9409, | |
"step": 690 | |
}, | |
{ | |
"epoch": 0.37, | |
"grad_norm": 44.70013427734375, | |
"learning_rate": 4.377777777777778e-05, | |
"loss": 0.7088, | |
"step": 700 | |
}, | |
{ | |
"epoch": 0.37, | |
"eval_accuracy": 0.7975, | |
"eval_f1_macro": 0.6808312841147881, | |
"eval_f1_micro": 0.7975, | |
"eval_loss": 0.7904818654060364, | |
"eval_runtime": 124.9315, | |
"eval_samples_per_second": 80.044, | |
"eval_steps_per_second": 2.505, | |
"step": 700 | |
}, | |
{ | |
"epoch": 0.38, | |
"grad_norm": 23.76036262512207, | |
"learning_rate": 4.368888888888889e-05, | |
"loss": 0.5632, | |
"step": 710 | |
}, | |
{ | |
"epoch": 0.38, | |
"grad_norm": 37.98194122314453, | |
"learning_rate": 4.36e-05, | |
"loss": 0.7349, | |
"step": 720 | |
}, | |
{ | |
"epoch": 0.39, | |
"grad_norm": 43.77289581298828, | |
"learning_rate": 4.351111111111111e-05, | |
"loss": 0.7132, | |
"step": 730 | |
}, | |
{ | |
"epoch": 0.39, | |
"grad_norm": 38.53476333618164, | |
"learning_rate": 4.3422222222222224e-05, | |
"loss": 0.6374, | |
"step": 740 | |
}, | |
{ | |
"epoch": 0.4, | |
"grad_norm": 36.89138412475586, | |
"learning_rate": 4.3333333333333334e-05, | |
"loss": 0.7341, | |
"step": 750 | |
}, | |
{ | |
"epoch": 0.41, | |
"grad_norm": 39.2047233581543, | |
"learning_rate": 4.324444444444445e-05, | |
"loss": 0.8566, | |
"step": 760 | |
}, | |
{ | |
"epoch": 0.41, | |
"grad_norm": 37.043670654296875, | |
"learning_rate": 4.315555555555556e-05, | |
"loss": 0.7392, | |
"step": 770 | |
}, | |
{ | |
"epoch": 0.42, | |
"grad_norm": 35.77628707885742, | |
"learning_rate": 4.3066666666666665e-05, | |
"loss": 0.7401, | |
"step": 780 | |
}, | |
{ | |
"epoch": 0.42, | |
"grad_norm": 32.939632415771484, | |
"learning_rate": 4.2977777777777776e-05, | |
"loss": 0.7112, | |
"step": 790 | |
}, | |
{ | |
"epoch": 0.43, | |
"grad_norm": 14.420007705688477, | |
"learning_rate": 4.2888888888888886e-05, | |
"loss": 0.6096, | |
"step": 800 | |
}, | |
{ | |
"epoch": 0.43, | |
"eval_accuracy": 0.8009, | |
"eval_f1_macro": 0.6822606117129679, | |
"eval_f1_micro": 0.8009, | |
"eval_loss": 0.7862304449081421, | |
"eval_runtime": 124.8581, | |
"eval_samples_per_second": 80.091, | |
"eval_steps_per_second": 2.507, | |
"step": 800 | |
}, | |
{ | |
"epoch": 0.43, | |
"grad_norm": 42.69435119628906, | |
"learning_rate": 4.2800000000000004e-05, | |
"loss": 0.6954, | |
"step": 810 | |
}, | |
{ | |
"epoch": 0.44, | |
"grad_norm": 45.64569091796875, | |
"learning_rate": 4.2711111111111114e-05, | |
"loss": 0.8801, | |
"step": 820 | |
}, | |
{ | |
"epoch": 0.44, | |
"grad_norm": 27.12781524658203, | |
"learning_rate": 4.2622222222222224e-05, | |
"loss": 0.8576, | |
"step": 830 | |
}, | |
{ | |
"epoch": 0.45, | |
"grad_norm": 48.820552825927734, | |
"learning_rate": 4.2533333333333335e-05, | |
"loss": 0.6565, | |
"step": 840 | |
}, | |
{ | |
"epoch": 0.45, | |
"grad_norm": 43.27635192871094, | |
"learning_rate": 4.2444444444444445e-05, | |
"loss": 0.7962, | |
"step": 850 | |
}, | |
{ | |
"epoch": 0.46, | |
"grad_norm": 39.85664367675781, | |
"learning_rate": 4.235555555555556e-05, | |
"loss": 0.9294, | |
"step": 860 | |
}, | |
{ | |
"epoch": 0.46, | |
"grad_norm": 45.196571350097656, | |
"learning_rate": 4.226666666666667e-05, | |
"loss": 0.9026, | |
"step": 870 | |
}, | |
{ | |
"epoch": 0.47, | |
"grad_norm": 29.43784523010254, | |
"learning_rate": 4.217777777777778e-05, | |
"loss": 0.7298, | |
"step": 880 | |
}, | |
{ | |
"epoch": 0.47, | |
"grad_norm": 36.50236129760742, | |
"learning_rate": 4.208888888888889e-05, | |
"loss": 0.787, | |
"step": 890 | |
}, | |
{ | |
"epoch": 0.48, | |
"grad_norm": 47.03664016723633, | |
"learning_rate": 4.2e-05, | |
"loss": 0.8682, | |
"step": 900 | |
}, | |
{ | |
"epoch": 0.48, | |
"eval_accuracy": 0.7987, | |
"eval_f1_macro": 0.6966502426266987, | |
"eval_f1_micro": 0.7987, | |
"eval_loss": 0.7767773270606995, | |
"eval_runtime": 125.0393, | |
"eval_samples_per_second": 79.975, | |
"eval_steps_per_second": 2.503, | |
"step": 900 | |
}, | |
{ | |
"epoch": 0.49, | |
"grad_norm": 39.72060012817383, | |
"learning_rate": 4.1911111111111115e-05, | |
"loss": 0.8498, | |
"step": 910 | |
}, | |
{ | |
"epoch": 0.49, | |
"grad_norm": 37.37692642211914, | |
"learning_rate": 4.1822222222222225e-05, | |
"loss": 0.718, | |
"step": 920 | |
}, | |
{ | |
"epoch": 0.5, | |
"grad_norm": 28.36815071105957, | |
"learning_rate": 4.1733333333333336e-05, | |
"loss": 0.6923, | |
"step": 930 | |
}, | |
{ | |
"epoch": 0.5, | |
"grad_norm": 41.33854675292969, | |
"learning_rate": 4.1644444444444446e-05, | |
"loss": 0.7234, | |
"step": 940 | |
}, | |
{ | |
"epoch": 0.51, | |
"grad_norm": 37.55774688720703, | |
"learning_rate": 4.155555555555556e-05, | |
"loss": 0.8102, | |
"step": 950 | |
}, | |
{ | |
"epoch": 0.51, | |
"grad_norm": 32.01072692871094, | |
"learning_rate": 4.146666666666667e-05, | |
"loss": 0.7338, | |
"step": 960 | |
}, | |
{ | |
"epoch": 0.52, | |
"grad_norm": 23.995277404785156, | |
"learning_rate": 4.1377777777777784e-05, | |
"loss": 0.7768, | |
"step": 970 | |
}, | |
{ | |
"epoch": 0.52, | |
"grad_norm": 37.772430419921875, | |
"learning_rate": 4.1288888888888895e-05, | |
"loss": 0.746, | |
"step": 980 | |
}, | |
{ | |
"epoch": 0.53, | |
"grad_norm": 45.859737396240234, | |
"learning_rate": 4.12e-05, | |
"loss": 0.7789, | |
"step": 990 | |
}, | |
{ | |
"epoch": 0.53, | |
"grad_norm": 41.380367279052734, | |
"learning_rate": 4.111111111111111e-05, | |
"loss": 0.6772, | |
"step": 1000 | |
}, | |
{ | |
"epoch": 0.53, | |
"eval_accuracy": 0.8094, | |
"eval_f1_macro": 0.6933975962115336, | |
"eval_f1_micro": 0.8094, | |
"eval_loss": 0.7299655079841614, | |
"eval_runtime": 125.167, | |
"eval_samples_per_second": 79.893, | |
"eval_steps_per_second": 2.501, | |
"step": 1000 | |
}, | |
{ | |
"epoch": 0.54, | |
"grad_norm": 29.1951904296875, | |
"learning_rate": 4.1022222222222226e-05, | |
"loss": 0.6658, | |
"step": 1010 | |
}, | |
{ | |
"epoch": 0.54, | |
"grad_norm": 32.311279296875, | |
"learning_rate": 4.093333333333334e-05, | |
"loss": 0.7022, | |
"step": 1020 | |
}, | |
{ | |
"epoch": 0.55, | |
"grad_norm": 30.058263778686523, | |
"learning_rate": 4.084444444444445e-05, | |
"loss": 0.8125, | |
"step": 1030 | |
}, | |
{ | |
"epoch": 0.55, | |
"grad_norm": 60.061588287353516, | |
"learning_rate": 4.075555555555556e-05, | |
"loss": 0.7732, | |
"step": 1040 | |
}, | |
{ | |
"epoch": 0.56, | |
"grad_norm": 40.17058181762695, | |
"learning_rate": 4.066666666666667e-05, | |
"loss": 0.6829, | |
"step": 1050 | |
}, | |
{ | |
"epoch": 0.57, | |
"grad_norm": 30.274438858032227, | |
"learning_rate": 4.057777777777778e-05, | |
"loss": 0.6306, | |
"step": 1060 | |
}, | |
{ | |
"epoch": 0.57, | |
"grad_norm": 32.10597229003906, | |
"learning_rate": 4.0488888888888896e-05, | |
"loss": 0.723, | |
"step": 1070 | |
}, | |
{ | |
"epoch": 0.58, | |
"grad_norm": 24.114013671875, | |
"learning_rate": 4.0400000000000006e-05, | |
"loss": 0.5363, | |
"step": 1080 | |
}, | |
{ | |
"epoch": 0.58, | |
"grad_norm": 34.77803421020508, | |
"learning_rate": 4.031111111111111e-05, | |
"loss": 0.6772, | |
"step": 1090 | |
}, | |
{ | |
"epoch": 0.59, | |
"grad_norm": 18.767980575561523, | |
"learning_rate": 4.022222222222222e-05, | |
"loss": 0.6224, | |
"step": 1100 | |
}, | |
{ | |
"epoch": 0.59, | |
"eval_accuracy": 0.8146, | |
"eval_f1_macro": 0.7190013669213452, | |
"eval_f1_micro": 0.8146, | |
"eval_loss": 0.675977349281311, | |
"eval_runtime": 125.1012, | |
"eval_samples_per_second": 79.935, | |
"eval_steps_per_second": 2.502, | |
"step": 1100 | |
}, | |
{ | |
"epoch": 0.59, | |
"grad_norm": 29.401121139526367, | |
"learning_rate": 4.013333333333333e-05, | |
"loss": 0.5357, | |
"step": 1110 | |
}, | |
{ | |
"epoch": 0.6, | |
"grad_norm": 34.86585235595703, | |
"learning_rate": 4.004444444444445e-05, | |
"loss": 0.7042, | |
"step": 1120 | |
}, | |
{ | |
"epoch": 0.6, | |
"grad_norm": 45.15092849731445, | |
"learning_rate": 3.995555555555556e-05, | |
"loss": 0.5536, | |
"step": 1130 | |
}, | |
{ | |
"epoch": 0.61, | |
"grad_norm": 49.75371551513672, | |
"learning_rate": 3.986666666666667e-05, | |
"loss": 0.7, | |
"step": 1140 | |
}, | |
{ | |
"epoch": 0.61, | |
"grad_norm": 35.728458404541016, | |
"learning_rate": 3.977777777777778e-05, | |
"loss": 0.6518, | |
"step": 1150 | |
}, | |
{ | |
"epoch": 0.62, | |
"grad_norm": 35.763145446777344, | |
"learning_rate": 3.968888888888889e-05, | |
"loss": 0.5926, | |
"step": 1160 | |
}, | |
{ | |
"epoch": 0.62, | |
"grad_norm": 34.241641998291016, | |
"learning_rate": 3.960000000000001e-05, | |
"loss": 0.7443, | |
"step": 1170 | |
}, | |
{ | |
"epoch": 0.63, | |
"grad_norm": 35.822608947753906, | |
"learning_rate": 3.951111111111112e-05, | |
"loss": 0.7653, | |
"step": 1180 | |
}, | |
{ | |
"epoch": 0.63, | |
"grad_norm": 37.08557891845703, | |
"learning_rate": 3.942222222222222e-05, | |
"loss": 0.6447, | |
"step": 1190 | |
}, | |
{ | |
"epoch": 0.64, | |
"grad_norm": 40.616458892822266, | |
"learning_rate": 3.933333333333333e-05, | |
"loss": 0.5875, | |
"step": 1200 | |
}, | |
{ | |
"epoch": 0.64, | |
"eval_accuracy": 0.8253, | |
"eval_f1_macro": 0.7442346170146863, | |
"eval_f1_micro": 0.8253, | |
"eval_loss": 0.6448789238929749, | |
"eval_runtime": 125.1421, | |
"eval_samples_per_second": 79.909, | |
"eval_steps_per_second": 2.501, | |
"step": 1200 | |
}, | |
{ | |
"epoch": 0.65, | |
"grad_norm": 43.901790618896484, | |
"learning_rate": 3.924444444444444e-05, | |
"loss": 0.6233, | |
"step": 1210 | |
}, | |
{ | |
"epoch": 0.65, | |
"grad_norm": 28.500553131103516, | |
"learning_rate": 3.915555555555556e-05, | |
"loss": 0.7282, | |
"step": 1220 | |
}, | |
{ | |
"epoch": 0.66, | |
"grad_norm": 41.72186279296875, | |
"learning_rate": 3.906666666666667e-05, | |
"loss": 0.6711, | |
"step": 1230 | |
}, | |
{ | |
"epoch": 0.66, | |
"grad_norm": 46.095245361328125, | |
"learning_rate": 3.897777777777778e-05, | |
"loss": 0.5736, | |
"step": 1240 | |
}, | |
{ | |
"epoch": 0.67, | |
"grad_norm": 41.30305862426758, | |
"learning_rate": 3.888888888888889e-05, | |
"loss": 0.6972, | |
"step": 1250 | |
}, | |
{ | |
"epoch": 0.67, | |
"grad_norm": 33.491661071777344, | |
"learning_rate": 3.88e-05, | |
"loss": 0.7147, | |
"step": 1260 | |
}, | |
{ | |
"epoch": 0.68, | |
"grad_norm": 40.78701400756836, | |
"learning_rate": 3.871111111111111e-05, | |
"loss": 0.5984, | |
"step": 1270 | |
}, | |
{ | |
"epoch": 0.68, | |
"grad_norm": 36.48966598510742, | |
"learning_rate": 3.862222222222223e-05, | |
"loss": 0.7407, | |
"step": 1280 | |
}, | |
{ | |
"epoch": 0.69, | |
"grad_norm": 39.34754180908203, | |
"learning_rate": 3.853333333333334e-05, | |
"loss": 0.5544, | |
"step": 1290 | |
}, | |
{ | |
"epoch": 0.69, | |
"grad_norm": 28.790693283081055, | |
"learning_rate": 3.844444444444444e-05, | |
"loss": 0.6147, | |
"step": 1300 | |
}, | |
{ | |
"epoch": 0.69, | |
"eval_accuracy": 0.8305, | |
"eval_f1_macro": 0.7207971849621393, | |
"eval_f1_micro": 0.8305, | |
"eval_loss": 0.6603367328643799, | |
"eval_runtime": 125.1561, | |
"eval_samples_per_second": 79.9, | |
"eval_steps_per_second": 2.501, | |
"step": 1300 | |
}, | |
{ | |
"epoch": 0.7, | |
"grad_norm": 23.91292381286621, | |
"learning_rate": 3.8355555555555553e-05, | |
"loss": 0.5826, | |
"step": 1310 | |
}, | |
{ | |
"epoch": 0.7, | |
"grad_norm": 39.718162536621094, | |
"learning_rate": 3.8266666666666664e-05, | |
"loss": 0.7496, | |
"step": 1320 | |
}, | |
{ | |
"epoch": 0.71, | |
"grad_norm": 28.492454528808594, | |
"learning_rate": 3.817777777777778e-05, | |
"loss": 0.611, | |
"step": 1330 | |
}, | |
{ | |
"epoch": 0.71, | |
"grad_norm": 34.915157318115234, | |
"learning_rate": 3.808888888888889e-05, | |
"loss": 0.7782, | |
"step": 1340 | |
}, | |
{ | |
"epoch": 0.72, | |
"grad_norm": 29.332368850708008, | |
"learning_rate": 3.8e-05, | |
"loss": 0.5906, | |
"step": 1350 | |
}, | |
{ | |
"epoch": 0.73, | |
"grad_norm": 25.36094856262207, | |
"learning_rate": 3.791111111111111e-05, | |
"loss": 0.5216, | |
"step": 1360 | |
}, | |
{ | |
"epoch": 0.73, | |
"grad_norm": 46.5117301940918, | |
"learning_rate": 3.782222222222222e-05, | |
"loss": 0.5596, | |
"step": 1370 | |
}, | |
{ | |
"epoch": 0.74, | |
"grad_norm": 35.714317321777344, | |
"learning_rate": 3.773333333333334e-05, | |
"loss": 0.6629, | |
"step": 1380 | |
}, | |
{ | |
"epoch": 0.74, | |
"grad_norm": 23.61018180847168, | |
"learning_rate": 3.764444444444445e-05, | |
"loss": 0.6112, | |
"step": 1390 | |
}, | |
{ | |
"epoch": 0.75, | |
"grad_norm": 26.34315299987793, | |
"learning_rate": 3.7555555555555554e-05, | |
"loss": 0.6355, | |
"step": 1400 | |
}, | |
{ | |
"epoch": 0.75, | |
"eval_accuracy": 0.8285, | |
"eval_f1_macro": 0.7294061203799771, | |
"eval_f1_micro": 0.8285, | |
"eval_loss": 0.6256009936332703, | |
"eval_runtime": 125.091, | |
"eval_samples_per_second": 79.942, | |
"eval_steps_per_second": 2.502, | |
"step": 1400 | |
}, | |
{ | |
"epoch": 0.75, | |
"grad_norm": 30.554851531982422, | |
"learning_rate": 3.7466666666666665e-05, | |
"loss": 0.6454, | |
"step": 1410 | |
}, | |
{ | |
"epoch": 0.76, | |
"grad_norm": 34.99703598022461, | |
"learning_rate": 3.7377777777777775e-05, | |
"loss": 0.5143, | |
"step": 1420 | |
}, | |
{ | |
"epoch": 0.76, | |
"grad_norm": 32.562931060791016, | |
"learning_rate": 3.728888888888889e-05, | |
"loss": 0.5926, | |
"step": 1430 | |
}, | |
{ | |
"epoch": 0.77, | |
"grad_norm": 35.89056396484375, | |
"learning_rate": 3.72e-05, | |
"loss": 0.5866, | |
"step": 1440 | |
}, | |
{ | |
"epoch": 0.77, | |
"grad_norm": 46.53374099731445, | |
"learning_rate": 3.7111111111111113e-05, | |
"loss": 0.6171, | |
"step": 1450 | |
}, | |
{ | |
"epoch": 0.78, | |
"grad_norm": 45.57334518432617, | |
"learning_rate": 3.7022222222222224e-05, | |
"loss": 0.6697, | |
"step": 1460 | |
}, | |
{ | |
"epoch": 0.78, | |
"grad_norm": 38.89950180053711, | |
"learning_rate": 3.6933333333333334e-05, | |
"loss": 0.7297, | |
"step": 1470 | |
}, | |
{ | |
"epoch": 0.79, | |
"grad_norm": 29.517240524291992, | |
"learning_rate": 3.6844444444444445e-05, | |
"loss": 0.6718, | |
"step": 1480 | |
}, | |
{ | |
"epoch": 0.79, | |
"grad_norm": 39.73881912231445, | |
"learning_rate": 3.675555555555556e-05, | |
"loss": 0.7189, | |
"step": 1490 | |
}, | |
{ | |
"epoch": 0.8, | |
"grad_norm": 32.796417236328125, | |
"learning_rate": 3.6666666666666666e-05, | |
"loss": 0.7076, | |
"step": 1500 | |
}, | |
{ | |
"epoch": 0.8, | |
"eval_accuracy": 0.8288, | |
"eval_f1_macro": 0.7290119185786764, | |
"eval_f1_micro": 0.8288, | |
"eval_loss": 0.6339758038520813, | |
"eval_runtime": 125.2044, | |
"eval_samples_per_second": 79.869, | |
"eval_steps_per_second": 2.5, | |
"step": 1500 | |
}, | |
{ | |
"epoch": 0.81, | |
"grad_norm": 46.705352783203125, | |
"learning_rate": 3.6577777777777776e-05, | |
"loss": 0.6376, | |
"step": 1510 | |
}, | |
{ | |
"epoch": 0.81, | |
"grad_norm": 37.39145278930664, | |
"learning_rate": 3.648888888888889e-05, | |
"loss": 0.6244, | |
"step": 1520 | |
}, | |
{ | |
"epoch": 0.82, | |
"grad_norm": 18.704368591308594, | |
"learning_rate": 3.6400000000000004e-05, | |
"loss": 0.5616, | |
"step": 1530 | |
}, | |
{ | |
"epoch": 0.82, | |
"grad_norm": 27.825958251953125, | |
"learning_rate": 3.6311111111111114e-05, | |
"loss": 0.5144, | |
"step": 1540 | |
}, | |
{ | |
"epoch": 0.83, | |
"grad_norm": 26.275487899780273, | |
"learning_rate": 3.6222222222222225e-05, | |
"loss": 0.703, | |
"step": 1550 | |
}, | |
{ | |
"epoch": 0.83, | |
"grad_norm": 39.88068771362305, | |
"learning_rate": 3.6133333333333335e-05, | |
"loss": 0.7873, | |
"step": 1560 | |
}, | |
{ | |
"epoch": 0.84, | |
"grad_norm": 37.45166015625, | |
"learning_rate": 3.6044444444444446e-05, | |
"loss": 0.6695, | |
"step": 1570 | |
}, | |
{ | |
"epoch": 0.84, | |
"grad_norm": 30.150066375732422, | |
"learning_rate": 3.5955555555555556e-05, | |
"loss": 0.6321, | |
"step": 1580 | |
}, | |
{ | |
"epoch": 0.85, | |
"grad_norm": 28.757957458496094, | |
"learning_rate": 3.586666666666667e-05, | |
"loss": 0.7031, | |
"step": 1590 | |
}, | |
{ | |
"epoch": 0.85, | |
"grad_norm": 18.669872283935547, | |
"learning_rate": 3.577777777777778e-05, | |
"loss": 0.4995, | |
"step": 1600 | |
}, | |
{ | |
"epoch": 0.85, | |
"eval_accuracy": 0.8315, | |
"eval_f1_macro": 0.7422265545358996, | |
"eval_f1_micro": 0.8315, | |
"eval_loss": 0.6185604333877563, | |
"eval_runtime": 125.1107, | |
"eval_samples_per_second": 79.929, | |
"eval_steps_per_second": 2.502, | |
"step": 1600 | |
}, | |
{ | |
"epoch": 0.86, | |
"grad_norm": 24.61341667175293, | |
"learning_rate": 3.568888888888889e-05, | |
"loss": 0.5862, | |
"step": 1610 | |
}, | |
{ | |
"epoch": 0.86, | |
"grad_norm": 39.579833984375, | |
"learning_rate": 3.56e-05, | |
"loss": 0.5583, | |
"step": 1620 | |
}, | |
{ | |
"epoch": 0.87, | |
"grad_norm": 35.16648864746094, | |
"learning_rate": 3.551111111111111e-05, | |
"loss": 0.6049, | |
"step": 1630 | |
}, | |
{ | |
"epoch": 0.87, | |
"grad_norm": 27.00229263305664, | |
"learning_rate": 3.5422222222222226e-05, | |
"loss": 0.7742, | |
"step": 1640 | |
}, | |
{ | |
"epoch": 0.88, | |
"grad_norm": 41.69742202758789, | |
"learning_rate": 3.5333333333333336e-05, | |
"loss": 0.754, | |
"step": 1650 | |
}, | |
{ | |
"epoch": 0.89, | |
"grad_norm": 39.063114166259766, | |
"learning_rate": 3.5244444444444447e-05, | |
"loss": 0.5125, | |
"step": 1660 | |
}, | |
{ | |
"epoch": 0.89, | |
"grad_norm": 36.81621170043945, | |
"learning_rate": 3.515555555555556e-05, | |
"loss": 0.5039, | |
"step": 1670 | |
}, | |
{ | |
"epoch": 0.9, | |
"grad_norm": 43.19419860839844, | |
"learning_rate": 3.506666666666667e-05, | |
"loss": 0.6167, | |
"step": 1680 | |
}, | |
{ | |
"epoch": 0.9, | |
"grad_norm": 36.92513656616211, | |
"learning_rate": 3.4977777777777785e-05, | |
"loss": 0.5754, | |
"step": 1690 | |
}, | |
{ | |
"epoch": 0.91, | |
"grad_norm": 35.30787658691406, | |
"learning_rate": 3.4888888888888895e-05, | |
"loss": 0.5754, | |
"step": 1700 | |
}, | |
{ | |
"epoch": 0.91, | |
"eval_accuracy": 0.8402, | |
"eval_f1_macro": 0.748230160429884, | |
"eval_f1_micro": 0.8402, | |
"eval_loss": 0.610479474067688, | |
"eval_runtime": 125.3127, | |
"eval_samples_per_second": 79.8, | |
"eval_steps_per_second": 2.498, | |
"step": 1700 | |
}, | |
{ | |
"epoch": 0.91, | |
"grad_norm": 29.194902420043945, | |
"learning_rate": 3.48e-05, | |
"loss": 0.6592, | |
"step": 1710 | |
}, | |
{ | |
"epoch": 0.92, | |
"grad_norm": 45.93604278564453, | |
"learning_rate": 3.471111111111111e-05, | |
"loss": 0.5685, | |
"step": 1720 | |
}, | |
{ | |
"epoch": 0.92, | |
"grad_norm": 29.713573455810547, | |
"learning_rate": 3.462222222222222e-05, | |
"loss": 0.6759, | |
"step": 1730 | |
}, | |
{ | |
"epoch": 0.93, | |
"grad_norm": 30.30419921875, | |
"learning_rate": 3.453333333333334e-05, | |
"loss": 0.5521, | |
"step": 1740 | |
}, | |
{ | |
"epoch": 0.93, | |
"grad_norm": 32.098819732666016, | |
"learning_rate": 3.444444444444445e-05, | |
"loss": 0.5322, | |
"step": 1750 | |
}, | |
{ | |
"epoch": 0.94, | |
"grad_norm": 36.17616271972656, | |
"learning_rate": 3.435555555555556e-05, | |
"loss": 0.5412, | |
"step": 1760 | |
}, | |
{ | |
"epoch": 0.94, | |
"grad_norm": 31.130359649658203, | |
"learning_rate": 3.426666666666667e-05, | |
"loss": 0.4915, | |
"step": 1770 | |
}, | |
{ | |
"epoch": 0.95, | |
"grad_norm": 29.99827003479004, | |
"learning_rate": 3.417777777777778e-05, | |
"loss": 0.5745, | |
"step": 1780 | |
}, | |
{ | |
"epoch": 0.95, | |
"grad_norm": 34.081581115722656, | |
"learning_rate": 3.408888888888889e-05, | |
"loss": 0.4867, | |
"step": 1790 | |
}, | |
{ | |
"epoch": 0.96, | |
"grad_norm": 27.07353401184082, | |
"learning_rate": 3.4000000000000007e-05, | |
"loss": 0.6775, | |
"step": 1800 | |
}, | |
{ | |
"epoch": 0.96, | |
"eval_accuracy": 0.8369, | |
"eval_f1_macro": 0.7531105549201089, | |
"eval_f1_micro": 0.8369, | |
"eval_loss": 0.5946730375289917, | |
"eval_runtime": 125.1987, | |
"eval_samples_per_second": 79.873, | |
"eval_steps_per_second": 2.5, | |
"step": 1800 | |
}, | |
{ | |
"epoch": 0.97, | |
"grad_norm": 30.427448272705078, | |
"learning_rate": 3.391111111111111e-05, | |
"loss": 0.6234, | |
"step": 1810 | |
}, | |
{ | |
"epoch": 0.97, | |
"grad_norm": 24.822948455810547, | |
"learning_rate": 3.382222222222222e-05, | |
"loss": 0.761, | |
"step": 1820 | |
}, | |
{ | |
"epoch": 0.98, | |
"grad_norm": 20.0987606048584, | |
"learning_rate": 3.373333333333333e-05, | |
"loss": 0.5981, | |
"step": 1830 | |
}, | |
{ | |
"epoch": 0.98, | |
"grad_norm": 29.777557373046875, | |
"learning_rate": 3.364444444444445e-05, | |
"loss": 0.5722, | |
"step": 1840 | |
}, | |
{ | |
"epoch": 0.99, | |
"grad_norm": 36.3715705871582, | |
"learning_rate": 3.355555555555556e-05, | |
"loss": 0.5465, | |
"step": 1850 | |
}, | |
{ | |
"epoch": 0.99, | |
"grad_norm": 36.25810623168945, | |
"learning_rate": 3.346666666666667e-05, | |
"loss": 0.5046, | |
"step": 1860 | |
}, | |
{ | |
"epoch": 1.0, | |
"grad_norm": 40.49335479736328, | |
"learning_rate": 3.337777777777778e-05, | |
"loss": 0.6467, | |
"step": 1870 | |
}, | |
{ | |
"epoch": 1.0, | |
"grad_norm": 24.9757022857666, | |
"learning_rate": 3.328888888888889e-05, | |
"loss": 0.5147, | |
"step": 1880 | |
}, | |
{ | |
"epoch": 1.01, | |
"grad_norm": 25.19569969177246, | |
"learning_rate": 3.32e-05, | |
"loss": 0.4116, | |
"step": 1890 | |
}, | |
{ | |
"epoch": 1.01, | |
"grad_norm": 18.321523666381836, | |
"learning_rate": 3.311111111111112e-05, | |
"loss": 0.3267, | |
"step": 1900 | |
}, | |
{ | |
"epoch": 1.01, | |
"eval_accuracy": 0.8528, | |
"eval_f1_macro": 0.7703652931864461, | |
"eval_f1_micro": 0.8528, | |
"eval_loss": 0.5678077936172485, | |
"eval_runtime": 125.1344, | |
"eval_samples_per_second": 79.914, | |
"eval_steps_per_second": 2.501, | |
"step": 1900 | |
}, | |
{ | |
"epoch": 1.02, | |
"grad_norm": 19.32207489013672, | |
"learning_rate": 3.302222222222222e-05, | |
"loss": 0.3264, | |
"step": 1910 | |
}, | |
{ | |
"epoch": 1.02, | |
"grad_norm": 27.11937141418457, | |
"learning_rate": 3.293333333333333e-05, | |
"loss": 0.3292, | |
"step": 1920 | |
}, | |
{ | |
"epoch": 1.03, | |
"grad_norm": 12.001670837402344, | |
"learning_rate": 3.284444444444444e-05, | |
"loss": 0.2931, | |
"step": 1930 | |
}, | |
{ | |
"epoch": 1.03, | |
"grad_norm": 43.393165588378906, | |
"learning_rate": 3.275555555555555e-05, | |
"loss": 0.3901, | |
"step": 1940 | |
}, | |
{ | |
"epoch": 1.04, | |
"grad_norm": 29.158109664916992, | |
"learning_rate": 3.266666666666667e-05, | |
"loss": 0.3262, | |
"step": 1950 | |
}, | |
{ | |
"epoch": 1.05, | |
"grad_norm": 26.838531494140625, | |
"learning_rate": 3.257777777777778e-05, | |
"loss": 0.337, | |
"step": 1960 | |
}, | |
{ | |
"epoch": 1.05, | |
"grad_norm": 32.79772186279297, | |
"learning_rate": 3.248888888888889e-05, | |
"loss": 0.312, | |
"step": 1970 | |
}, | |
{ | |
"epoch": 1.06, | |
"grad_norm": 18.537200927734375, | |
"learning_rate": 3.24e-05, | |
"loss": 0.2918, | |
"step": 1980 | |
}, | |
{ | |
"epoch": 1.06, | |
"grad_norm": 21.449663162231445, | |
"learning_rate": 3.231111111111111e-05, | |
"loss": 0.2633, | |
"step": 1990 | |
}, | |
{ | |
"epoch": 1.07, | |
"grad_norm": 21.72381591796875, | |
"learning_rate": 3.222222222222223e-05, | |
"loss": 0.2022, | |
"step": 2000 | |
}, | |
{ | |
"epoch": 1.07, | |
"eval_accuracy": 0.844, | |
"eval_f1_macro": 0.7638807977574182, | |
"eval_f1_micro": 0.844, | |
"eval_loss": 0.6360877752304077, | |
"eval_runtime": 125.2286, | |
"eval_samples_per_second": 79.854, | |
"eval_steps_per_second": 2.499, | |
"step": 2000 | |
}, | |
{ | |
"epoch": 1.07, | |
"grad_norm": 27.054370880126953, | |
"learning_rate": 3.213333333333334e-05, | |
"loss": 0.4437, | |
"step": 2010 | |
}, | |
{ | |
"epoch": 1.08, | |
"grad_norm": 22.364294052124023, | |
"learning_rate": 3.204444444444444e-05, | |
"loss": 0.2647, | |
"step": 2020 | |
}, | |
{ | |
"epoch": 1.08, | |
"grad_norm": 25.677640914916992, | |
"learning_rate": 3.1955555555555554e-05, | |
"loss": 0.3455, | |
"step": 2030 | |
}, | |
{ | |
"epoch": 1.09, | |
"grad_norm": 12.445222854614258, | |
"learning_rate": 3.1866666666666664e-05, | |
"loss": 0.2633, | |
"step": 2040 | |
}, | |
{ | |
"epoch": 1.09, | |
"grad_norm": 27.752634048461914, | |
"learning_rate": 3.177777777777778e-05, | |
"loss": 0.3091, | |
"step": 2050 | |
}, | |
{ | |
"epoch": 1.1, | |
"grad_norm": 34.70381546020508, | |
"learning_rate": 3.168888888888889e-05, | |
"loss": 0.3414, | |
"step": 2060 | |
}, | |
{ | |
"epoch": 1.1, | |
"grad_norm": 33.72677993774414, | |
"learning_rate": 3.16e-05, | |
"loss": 0.3628, | |
"step": 2070 | |
}, | |
{ | |
"epoch": 1.11, | |
"grad_norm": 19.364152908325195, | |
"learning_rate": 3.151111111111111e-05, | |
"loss": 0.3072, | |
"step": 2080 | |
}, | |
{ | |
"epoch": 1.11, | |
"grad_norm": 30.036209106445312, | |
"learning_rate": 3.142222222222222e-05, | |
"loss": 0.3389, | |
"step": 2090 | |
}, | |
{ | |
"epoch": 1.12, | |
"grad_norm": 19.264528274536133, | |
"learning_rate": 3.1333333333333334e-05, | |
"loss": 0.3831, | |
"step": 2100 | |
}, | |
{ | |
"epoch": 1.12, | |
"eval_accuracy": 0.8503, | |
"eval_f1_macro": 0.767154964941313, | |
"eval_f1_micro": 0.8503, | |
"eval_loss": 0.5956984162330627, | |
"eval_runtime": 125.0998, | |
"eval_samples_per_second": 79.936, | |
"eval_steps_per_second": 2.502, | |
"step": 2100 | |
}, | |
{ | |
"epoch": 1.13, | |
"grad_norm": 19.570907592773438, | |
"learning_rate": 3.124444444444445e-05, | |
"loss": 0.3448, | |
"step": 2110 | |
}, | |
{ | |
"epoch": 1.13, | |
"grad_norm": 30.030786514282227, | |
"learning_rate": 3.1155555555555555e-05, | |
"loss": 0.3049, | |
"step": 2120 | |
}, | |
{ | |
"epoch": 1.14, | |
"grad_norm": 25.751008987426758, | |
"learning_rate": 3.1066666666666665e-05, | |
"loss": 0.2604, | |
"step": 2130 | |
}, | |
{ | |
"epoch": 1.14, | |
"grad_norm": 29.36272430419922, | |
"learning_rate": 3.0977777777777776e-05, | |
"loss": 0.3661, | |
"step": 2140 | |
}, | |
{ | |
"epoch": 1.15, | |
"grad_norm": 33.18568801879883, | |
"learning_rate": 3.088888888888889e-05, | |
"loss": 0.2897, | |
"step": 2150 | |
}, | |
{ | |
"epoch": 1.15, | |
"grad_norm": 41.40974426269531, | |
"learning_rate": 3.08e-05, | |
"loss": 0.3758, | |
"step": 2160 | |
}, | |
{ | |
"epoch": 1.16, | |
"grad_norm": 21.188396453857422, | |
"learning_rate": 3.0711111111111114e-05, | |
"loss": 0.3537, | |
"step": 2170 | |
}, | |
{ | |
"epoch": 1.16, | |
"grad_norm": 20.362560272216797, | |
"learning_rate": 3.0622222222222224e-05, | |
"loss": 0.313, | |
"step": 2180 | |
}, | |
{ | |
"epoch": 1.17, | |
"grad_norm": 27.672611236572266, | |
"learning_rate": 3.0533333333333335e-05, | |
"loss": 0.3813, | |
"step": 2190 | |
}, | |
{ | |
"epoch": 1.17, | |
"grad_norm": 15.358328819274902, | |
"learning_rate": 3.044444444444445e-05, | |
"loss": 0.3235, | |
"step": 2200 | |
}, | |
{ | |
"epoch": 1.17, | |
"eval_accuracy": 0.8476, | |
"eval_f1_macro": 0.7685252040288097, | |
"eval_f1_micro": 0.8476, | |
"eval_loss": 0.6061830520629883, | |
"eval_runtime": 125.1594, | |
"eval_samples_per_second": 79.898, | |
"eval_steps_per_second": 2.501, | |
"step": 2200 | |
}, | |
{ | |
"epoch": 1.18, | |
"grad_norm": 20.829286575317383, | |
"learning_rate": 3.035555555555556e-05, | |
"loss": 0.4018, | |
"step": 2210 | |
}, | |
{ | |
"epoch": 1.18, | |
"grad_norm": 32.589256286621094, | |
"learning_rate": 3.0266666666666666e-05, | |
"loss": 0.4166, | |
"step": 2220 | |
}, | |
{ | |
"epoch": 1.19, | |
"grad_norm": 14.95271110534668, | |
"learning_rate": 3.0177777777777776e-05, | |
"loss": 0.3814, | |
"step": 2230 | |
}, | |
{ | |
"epoch": 1.19, | |
"grad_norm": 32.90339279174805, | |
"learning_rate": 3.008888888888889e-05, | |
"loss": 0.318, | |
"step": 2240 | |
}, | |
{ | |
"epoch": 1.2, | |
"grad_norm": 29.13666343688965, | |
"learning_rate": 3e-05, | |
"loss": 0.3152, | |
"step": 2250 | |
}, | |
{ | |
"epoch": 1.21, | |
"grad_norm": 20.47429847717285, | |
"learning_rate": 2.991111111111111e-05, | |
"loss": 0.4088, | |
"step": 2260 | |
}, | |
{ | |
"epoch": 1.21, | |
"grad_norm": 32.37903594970703, | |
"learning_rate": 2.9822222222222225e-05, | |
"loss": 0.3615, | |
"step": 2270 | |
}, | |
{ | |
"epoch": 1.22, | |
"grad_norm": 29.304582595825195, | |
"learning_rate": 2.9733333333333336e-05, | |
"loss": 0.2779, | |
"step": 2280 | |
}, | |
{ | |
"epoch": 1.22, | |
"grad_norm": 15.094827651977539, | |
"learning_rate": 2.9644444444444446e-05, | |
"loss": 0.2391, | |
"step": 2290 | |
}, | |
{ | |
"epoch": 1.23, | |
"grad_norm": 15.804471969604492, | |
"learning_rate": 2.955555555555556e-05, | |
"loss": 0.2279, | |
"step": 2300 | |
}, | |
{ | |
"epoch": 1.23, | |
"eval_accuracy": 0.847, | |
"eval_f1_macro": 0.7658476370782488, | |
"eval_f1_micro": 0.847, | |
"eval_loss": 0.6254826188087463, | |
"eval_runtime": 125.1512, | |
"eval_samples_per_second": 79.903, | |
"eval_steps_per_second": 2.501, | |
"step": 2300 | |
}, | |
{ | |
"epoch": 1.23, | |
"grad_norm": 12.60870361328125, | |
"learning_rate": 2.946666666666667e-05, | |
"loss": 0.4015, | |
"step": 2310 | |
}, | |
{ | |
"epoch": 1.24, | |
"grad_norm": 27.877355575561523, | |
"learning_rate": 2.937777777777778e-05, | |
"loss": 0.3093, | |
"step": 2320 | |
}, | |
{ | |
"epoch": 1.24, | |
"grad_norm": 15.59749698638916, | |
"learning_rate": 2.9288888888888888e-05, | |
"loss": 0.2629, | |
"step": 2330 | |
}, | |
{ | |
"epoch": 1.25, | |
"grad_norm": 28.317983627319336, | |
"learning_rate": 2.9199999999999998e-05, | |
"loss": 0.2197, | |
"step": 2340 | |
}, | |
{ | |
"epoch": 1.25, | |
"grad_norm": 27.348453521728516, | |
"learning_rate": 2.9111111111111112e-05, | |
"loss": 0.3047, | |
"step": 2350 | |
}, | |
{ | |
"epoch": 1.26, | |
"grad_norm": 17.648712158203125, | |
"learning_rate": 2.9022222222222223e-05, | |
"loss": 0.3159, | |
"step": 2360 | |
}, | |
{ | |
"epoch": 1.26, | |
"grad_norm": 20.40242576599121, | |
"learning_rate": 2.8933333333333333e-05, | |
"loss": 0.3453, | |
"step": 2370 | |
}, | |
{ | |
"epoch": 1.27, | |
"grad_norm": 24.179431915283203, | |
"learning_rate": 2.8844444444444447e-05, | |
"loss": 0.3431, | |
"step": 2380 | |
}, | |
{ | |
"epoch": 1.27, | |
"grad_norm": 32.673763275146484, | |
"learning_rate": 2.8755555555555557e-05, | |
"loss": 0.3027, | |
"step": 2390 | |
}, | |
{ | |
"epoch": 1.28, | |
"grad_norm": 28.0662841796875, | |
"learning_rate": 2.8666666666666668e-05, | |
"loss": 0.3224, | |
"step": 2400 | |
}, | |
{ | |
"epoch": 1.28, | |
"eval_accuracy": 0.8537, | |
"eval_f1_macro": 0.7772010038894511, | |
"eval_f1_micro": 0.8537, | |
"eval_loss": 0.5753965973854065, | |
"eval_runtime": 125.2268, | |
"eval_samples_per_second": 79.855, | |
"eval_steps_per_second": 2.499, | |
"step": 2400 | |
}, | |
{ | |
"epoch": 1.29, | |
"grad_norm": 20.53518295288086, | |
"learning_rate": 2.857777777777778e-05, | |
"loss": 0.2814, | |
"step": 2410 | |
}, | |
{ | |
"epoch": 1.29, | |
"grad_norm": 25.769210815429688, | |
"learning_rate": 2.8488888888888892e-05, | |
"loss": 0.369, | |
"step": 2420 | |
}, | |
{ | |
"epoch": 1.3, | |
"grad_norm": 13.653368949890137, | |
"learning_rate": 2.84e-05, | |
"loss": 0.2124, | |
"step": 2430 | |
}, | |
{ | |
"epoch": 1.3, | |
"grad_norm": 25.229494094848633, | |
"learning_rate": 2.831111111111111e-05, | |
"loss": 0.3308, | |
"step": 2440 | |
}, | |
{ | |
"epoch": 1.31, | |
"grad_norm": 27.03401756286621, | |
"learning_rate": 2.8222222222222223e-05, | |
"loss": 0.3702, | |
"step": 2450 | |
}, | |
{ | |
"epoch": 1.31, | |
"grad_norm": 28.564023971557617, | |
"learning_rate": 2.8133333333333334e-05, | |
"loss": 0.3641, | |
"step": 2460 | |
}, | |
{ | |
"epoch": 1.32, | |
"grad_norm": 30.772727966308594, | |
"learning_rate": 2.8044444444444444e-05, | |
"loss": 0.3255, | |
"step": 2470 | |
}, | |
{ | |
"epoch": 1.32, | |
"grad_norm": 33.72224426269531, | |
"learning_rate": 2.7955555555555558e-05, | |
"loss": 0.3417, | |
"step": 2480 | |
}, | |
{ | |
"epoch": 1.33, | |
"grad_norm": 6.024286270141602, | |
"learning_rate": 2.786666666666667e-05, | |
"loss": 0.2718, | |
"step": 2490 | |
}, | |
{ | |
"epoch": 1.33, | |
"grad_norm": 21.70349884033203, | |
"learning_rate": 2.777777777777778e-05, | |
"loss": 0.3281, | |
"step": 2500 | |
}, | |
{ | |
"epoch": 1.33, | |
"eval_accuracy": 0.8598, | |
"eval_f1_macro": 0.7768837492760268, | |
"eval_f1_micro": 0.8598, | |
"eval_loss": 0.5763229727745056, | |
"eval_runtime": 125.2018, | |
"eval_samples_per_second": 79.871, | |
"eval_steps_per_second": 2.5, | |
"step": 2500 | |
}, | |
{ | |
"epoch": 1.34, | |
"grad_norm": 24.865785598754883, | |
"learning_rate": 2.7688888888888893e-05, | |
"loss": 0.2607, | |
"step": 2510 | |
}, | |
{ | |
"epoch": 1.34, | |
"grad_norm": 21.175254821777344, | |
"learning_rate": 2.7600000000000003e-05, | |
"loss": 0.3339, | |
"step": 2520 | |
}, | |
{ | |
"epoch": 1.35, | |
"grad_norm": 23.783727645874023, | |
"learning_rate": 2.751111111111111e-05, | |
"loss": 0.261, | |
"step": 2530 | |
}, | |
{ | |
"epoch": 1.35, | |
"grad_norm": 28.94060707092285, | |
"learning_rate": 2.742222222222222e-05, | |
"loss": 0.3169, | |
"step": 2540 | |
}, | |
{ | |
"epoch": 1.36, | |
"grad_norm": 30.291284561157227, | |
"learning_rate": 2.733333333333333e-05, | |
"loss": 0.2878, | |
"step": 2550 | |
}, | |
{ | |
"epoch": 1.37, | |
"grad_norm": 26.320711135864258, | |
"learning_rate": 2.7244444444444445e-05, | |
"loss": 0.2883, | |
"step": 2560 | |
}, | |
{ | |
"epoch": 1.37, | |
"grad_norm": 35.11587905883789, | |
"learning_rate": 2.7155555555555556e-05, | |
"loss": 0.3464, | |
"step": 2570 | |
}, | |
{ | |
"epoch": 1.38, | |
"grad_norm": 21.037996292114258, | |
"learning_rate": 2.706666666666667e-05, | |
"loss": 0.3166, | |
"step": 2580 | |
}, | |
{ | |
"epoch": 1.38, | |
"grad_norm": 21.80624008178711, | |
"learning_rate": 2.697777777777778e-05, | |
"loss": 0.2901, | |
"step": 2590 | |
}, | |
{ | |
"epoch": 1.39, | |
"grad_norm": 33.16673278808594, | |
"learning_rate": 2.688888888888889e-05, | |
"loss": 0.3909, | |
"step": 2600 | |
}, | |
{ | |
"epoch": 1.39, | |
"eval_accuracy": 0.8545, | |
"eval_f1_macro": 0.7778240850537461, | |
"eval_f1_micro": 0.8545, | |
"eval_loss": 0.5518567562103271, | |
"eval_runtime": 125.2624, | |
"eval_samples_per_second": 79.832, | |
"eval_steps_per_second": 2.499, | |
"step": 2600 | |
}, | |
{ | |
"epoch": 1.39, | |
"grad_norm": 24.126609802246094, | |
"learning_rate": 2.6800000000000004e-05, | |
"loss": 0.3126, | |
"step": 2610 | |
}, | |
{ | |
"epoch": 1.4, | |
"grad_norm": 20.498868942260742, | |
"learning_rate": 2.6711111111111115e-05, | |
"loss": 0.2461, | |
"step": 2620 | |
}, | |
{ | |
"epoch": 1.4, | |
"grad_norm": 26.896896362304688, | |
"learning_rate": 2.6622222222222225e-05, | |
"loss": 0.2962, | |
"step": 2630 | |
}, | |
{ | |
"epoch": 1.41, | |
"grad_norm": 27.555248260498047, | |
"learning_rate": 2.6533333333333332e-05, | |
"loss": 0.3247, | |
"step": 2640 | |
}, | |
{ | |
"epoch": 1.41, | |
"grad_norm": 20.325254440307617, | |
"learning_rate": 2.6444444444444443e-05, | |
"loss": 0.3544, | |
"step": 2650 | |
}, | |
{ | |
"epoch": 1.42, | |
"grad_norm": 35.30345153808594, | |
"learning_rate": 2.6355555555555557e-05, | |
"loss": 0.355, | |
"step": 2660 | |
}, | |
{ | |
"epoch": 1.42, | |
"grad_norm": 19.69151496887207, | |
"learning_rate": 2.6266666666666667e-05, | |
"loss": 0.3337, | |
"step": 2670 | |
}, | |
{ | |
"epoch": 1.43, | |
"grad_norm": 21.770431518554688, | |
"learning_rate": 2.6177777777777777e-05, | |
"loss": 0.3063, | |
"step": 2680 | |
}, | |
{ | |
"epoch": 1.43, | |
"grad_norm": 20.847126007080078, | |
"learning_rate": 2.608888888888889e-05, | |
"loss": 0.27, | |
"step": 2690 | |
}, | |
{ | |
"epoch": 1.44, | |
"grad_norm": 31.529712677001953, | |
"learning_rate": 2.6000000000000002e-05, | |
"loss": 0.3064, | |
"step": 2700 | |
}, | |
{ | |
"epoch": 1.44, | |
"eval_accuracy": 0.8536, | |
"eval_f1_macro": 0.7790368774303406, | |
"eval_f1_micro": 0.8536, | |
"eval_loss": 0.5842456221580505, | |
"eval_runtime": 125.3409, | |
"eval_samples_per_second": 79.782, | |
"eval_steps_per_second": 2.497, | |
"step": 2700 | |
}, | |
{ | |
"epoch": 1.45, | |
"grad_norm": 39.8318977355957, | |
"learning_rate": 2.5911111111111112e-05, | |
"loss": 0.2705, | |
"step": 2710 | |
}, | |
{ | |
"epoch": 1.45, | |
"grad_norm": 29.740388870239258, | |
"learning_rate": 2.5822222222222226e-05, | |
"loss": 0.2761, | |
"step": 2720 | |
}, | |
{ | |
"epoch": 1.46, | |
"grad_norm": 33.67303466796875, | |
"learning_rate": 2.5733333333333337e-05, | |
"loss": 0.2903, | |
"step": 2730 | |
}, | |
{ | |
"epoch": 1.46, | |
"grad_norm": 18.60785484313965, | |
"learning_rate": 2.5644444444444444e-05, | |
"loss": 0.311, | |
"step": 2740 | |
}, | |
{ | |
"epoch": 1.47, | |
"grad_norm": 22.325334548950195, | |
"learning_rate": 2.5555555555555554e-05, | |
"loss": 0.3542, | |
"step": 2750 | |
}, | |
{ | |
"epoch": 1.47, | |
"grad_norm": 9.186369895935059, | |
"learning_rate": 2.5466666666666668e-05, | |
"loss": 0.2995, | |
"step": 2760 | |
}, | |
{ | |
"epoch": 1.48, | |
"grad_norm": 17.34896469116211, | |
"learning_rate": 2.537777777777778e-05, | |
"loss": 0.3195, | |
"step": 2770 | |
}, | |
{ | |
"epoch": 1.48, | |
"grad_norm": 33.827674865722656, | |
"learning_rate": 2.528888888888889e-05, | |
"loss": 0.3021, | |
"step": 2780 | |
}, | |
{ | |
"epoch": 1.49, | |
"grad_norm": 14.778377532958984, | |
"learning_rate": 2.5200000000000003e-05, | |
"loss": 0.2876, | |
"step": 2790 | |
}, | |
{ | |
"epoch": 1.49, | |
"grad_norm": 20.28546714782715, | |
"learning_rate": 2.5111111111111113e-05, | |
"loss": 0.2333, | |
"step": 2800 | |
}, | |
{ | |
"epoch": 1.49, | |
"eval_accuracy": 0.8447, | |
"eval_f1_macro": 0.7673871147650553, | |
"eval_f1_micro": 0.8447, | |
"eval_loss": 0.6084045767784119, | |
"eval_runtime": 125.269, | |
"eval_samples_per_second": 79.828, | |
"eval_steps_per_second": 2.499, | |
"step": 2800 | |
}, | |
{ | |
"epoch": 1.5, | |
"grad_norm": 26.260639190673828, | |
"learning_rate": 2.5022222222222224e-05, | |
"loss": 0.2639, | |
"step": 2810 | |
}, | |
{ | |
"epoch": 1.5, | |
"grad_norm": 22.30939292907715, | |
"learning_rate": 2.4933333333333334e-05, | |
"loss": 0.3508, | |
"step": 2820 | |
}, | |
{ | |
"epoch": 1.51, | |
"grad_norm": 29.393909454345703, | |
"learning_rate": 2.4844444444444444e-05, | |
"loss": 0.2569, | |
"step": 2830 | |
}, | |
{ | |
"epoch": 1.51, | |
"grad_norm": 26.341440200805664, | |
"learning_rate": 2.475555555555556e-05, | |
"loss": 0.2746, | |
"step": 2840 | |
}, | |
{ | |
"epoch": 1.52, | |
"grad_norm": 10.78348445892334, | |
"learning_rate": 2.466666666666667e-05, | |
"loss": 0.2224, | |
"step": 2850 | |
}, | |
{ | |
"epoch": 1.53, | |
"grad_norm": 23.771703720092773, | |
"learning_rate": 2.457777777777778e-05, | |
"loss": 0.2819, | |
"step": 2860 | |
}, | |
{ | |
"epoch": 1.53, | |
"grad_norm": 20.408435821533203, | |
"learning_rate": 2.448888888888889e-05, | |
"loss": 0.3146, | |
"step": 2870 | |
}, | |
{ | |
"epoch": 1.54, | |
"grad_norm": 18.23904800415039, | |
"learning_rate": 2.44e-05, | |
"loss": 0.3073, | |
"step": 2880 | |
}, | |
{ | |
"epoch": 1.54, | |
"grad_norm": 13.120834350585938, | |
"learning_rate": 2.431111111111111e-05, | |
"loss": 0.2746, | |
"step": 2890 | |
}, | |
{ | |
"epoch": 1.55, | |
"grad_norm": 26.703229904174805, | |
"learning_rate": 2.4222222222222224e-05, | |
"loss": 0.2361, | |
"step": 2900 | |
}, | |
{ | |
"epoch": 1.55, | |
"eval_accuracy": 0.8588, | |
"eval_f1_macro": 0.7853092454906397, | |
"eval_f1_micro": 0.8588, | |
"eval_loss": 0.5974697470664978, | |
"eval_runtime": 125.2009, | |
"eval_samples_per_second": 79.872, | |
"eval_steps_per_second": 2.5, | |
"step": 2900 | |
}, | |
{ | |
"epoch": 1.55, | |
"grad_norm": 22.55280876159668, | |
"learning_rate": 2.4133333333333335e-05, | |
"loss": 0.3899, | |
"step": 2910 | |
}, | |
{ | |
"epoch": 1.56, | |
"grad_norm": 18.248414993286133, | |
"learning_rate": 2.4044444444444445e-05, | |
"loss": 0.2995, | |
"step": 2920 | |
}, | |
{ | |
"epoch": 1.56, | |
"grad_norm": 26.95282554626465, | |
"learning_rate": 2.3955555555555556e-05, | |
"loss": 0.4188, | |
"step": 2930 | |
}, | |
{ | |
"epoch": 1.57, | |
"grad_norm": 18.83556365966797, | |
"learning_rate": 2.3866666666666666e-05, | |
"loss": 0.2215, | |
"step": 2940 | |
}, | |
{ | |
"epoch": 1.57, | |
"grad_norm": 21.471210479736328, | |
"learning_rate": 2.377777777777778e-05, | |
"loss": 0.3084, | |
"step": 2950 | |
}, | |
{ | |
"epoch": 1.58, | |
"grad_norm": 23.17391014099121, | |
"learning_rate": 2.368888888888889e-05, | |
"loss": 0.3674, | |
"step": 2960 | |
}, | |
{ | |
"epoch": 1.58, | |
"grad_norm": 31.083335876464844, | |
"learning_rate": 2.36e-05, | |
"loss": 0.286, | |
"step": 2970 | |
}, | |
{ | |
"epoch": 1.59, | |
"grad_norm": 27.996196746826172, | |
"learning_rate": 2.351111111111111e-05, | |
"loss": 0.3615, | |
"step": 2980 | |
}, | |
{ | |
"epoch": 1.59, | |
"grad_norm": 25.99715805053711, | |
"learning_rate": 2.3422222222222222e-05, | |
"loss": 0.2546, | |
"step": 2990 | |
}, | |
{ | |
"epoch": 1.6, | |
"grad_norm": 6.342403888702393, | |
"learning_rate": 2.3333333333333336e-05, | |
"loss": 0.3415, | |
"step": 3000 | |
}, | |
{ | |
"epoch": 1.6, | |
"eval_accuracy": 0.8572, | |
"eval_f1_macro": 0.7844301563007002, | |
"eval_f1_micro": 0.8572, | |
"eval_loss": 0.5700847506523132, | |
"eval_runtime": 125.2898, | |
"eval_samples_per_second": 79.815, | |
"eval_steps_per_second": 2.498, | |
"step": 3000 | |
}, | |
{ | |
"epoch": 1.61, | |
"grad_norm": 30.417648315429688, | |
"learning_rate": 2.3244444444444446e-05, | |
"loss": 0.3269, | |
"step": 3010 | |
}, | |
{ | |
"epoch": 1.61, | |
"grad_norm": 33.5994873046875, | |
"learning_rate": 2.3155555555555557e-05, | |
"loss": 0.342, | |
"step": 3020 | |
}, | |
{ | |
"epoch": 1.62, | |
"grad_norm": 30.278759002685547, | |
"learning_rate": 2.3066666666666667e-05, | |
"loss": 0.3424, | |
"step": 3030 | |
}, | |
{ | |
"epoch": 1.62, | |
"grad_norm": 28.47062873840332, | |
"learning_rate": 2.2977777777777778e-05, | |
"loss": 0.3254, | |
"step": 3040 | |
}, | |
{ | |
"epoch": 1.63, | |
"grad_norm": 27.17629051208496, | |
"learning_rate": 2.288888888888889e-05, | |
"loss": 0.4687, | |
"step": 3050 | |
}, | |
{ | |
"epoch": 1.63, | |
"grad_norm": 20.395219802856445, | |
"learning_rate": 2.2800000000000002e-05, | |
"loss": 0.2192, | |
"step": 3060 | |
}, | |
{ | |
"epoch": 1.64, | |
"grad_norm": 16.45476722717285, | |
"learning_rate": 2.2711111111111112e-05, | |
"loss": 0.5145, | |
"step": 3070 | |
}, | |
{ | |
"epoch": 1.64, | |
"grad_norm": 30.59330177307129, | |
"learning_rate": 2.2622222222222223e-05, | |
"loss": 0.2724, | |
"step": 3080 | |
}, | |
{ | |
"epoch": 1.65, | |
"grad_norm": 18.152570724487305, | |
"learning_rate": 2.2533333333333333e-05, | |
"loss": 0.319, | |
"step": 3090 | |
}, | |
{ | |
"epoch": 1.65, | |
"grad_norm": 26.22066879272461, | |
"learning_rate": 2.2444444444444447e-05, | |
"loss": 0.2535, | |
"step": 3100 | |
}, | |
{ | |
"epoch": 1.65, | |
"eval_accuracy": 0.8618, | |
"eval_f1_macro": 0.7827999074535285, | |
"eval_f1_micro": 0.8618, | |
"eval_loss": 0.5556691288948059, | |
"eval_runtime": 125.2171, | |
"eval_samples_per_second": 79.861, | |
"eval_steps_per_second": 2.5, | |
"step": 3100 | |
}, | |
{ | |
"epoch": 1.66, | |
"grad_norm": 12.443004608154297, | |
"learning_rate": 2.2355555555555558e-05, | |
"loss": 0.3178, | |
"step": 3110 | |
}, | |
{ | |
"epoch": 1.66, | |
"grad_norm": 27.130870819091797, | |
"learning_rate": 2.2266666666666668e-05, | |
"loss": 0.35, | |
"step": 3120 | |
}, | |
{ | |
"epoch": 1.67, | |
"grad_norm": 18.87975311279297, | |
"learning_rate": 2.217777777777778e-05, | |
"loss": 0.3123, | |
"step": 3130 | |
}, | |
{ | |
"epoch": 1.67, | |
"grad_norm": 25.946163177490234, | |
"learning_rate": 2.208888888888889e-05, | |
"loss": 0.3041, | |
"step": 3140 | |
}, | |
{ | |
"epoch": 1.68, | |
"grad_norm": 17.42608642578125, | |
"learning_rate": 2.2000000000000003e-05, | |
"loss": 0.2747, | |
"step": 3150 | |
}, | |
{ | |
"epoch": 1.69, | |
"grad_norm": 19.43211555480957, | |
"learning_rate": 2.1911111111111113e-05, | |
"loss": 0.2509, | |
"step": 3160 | |
}, | |
{ | |
"epoch": 1.69, | |
"grad_norm": 28.090959548950195, | |
"learning_rate": 2.1822222222222224e-05, | |
"loss": 0.2771, | |
"step": 3170 | |
}, | |
{ | |
"epoch": 1.7, | |
"grad_norm": 18.299936294555664, | |
"learning_rate": 2.1733333333333334e-05, | |
"loss": 0.3804, | |
"step": 3180 | |
}, | |
{ | |
"epoch": 1.7, | |
"grad_norm": 17.102720260620117, | |
"learning_rate": 2.1644444444444445e-05, | |
"loss": 0.2603, | |
"step": 3190 | |
}, | |
{ | |
"epoch": 1.71, | |
"grad_norm": 32.00794219970703, | |
"learning_rate": 2.1555555555555555e-05, | |
"loss": 0.2356, | |
"step": 3200 | |
}, | |
{ | |
"epoch": 1.71, | |
"eval_accuracy": 0.8612, | |
"eval_f1_macro": 0.782152177469759, | |
"eval_f1_micro": 0.8612, | |
"eval_loss": 0.5241914987564087, | |
"eval_runtime": 125.2745, | |
"eval_samples_per_second": 79.825, | |
"eval_steps_per_second": 2.499, | |
"step": 3200 | |
}, | |
{ | |
"epoch": 1.71, | |
"grad_norm": 31.098926544189453, | |
"learning_rate": 2.146666666666667e-05, | |
"loss": 0.257, | |
"step": 3210 | |
}, | |
{ | |
"epoch": 1.72, | |
"grad_norm": 22.57906723022461, | |
"learning_rate": 2.137777777777778e-05, | |
"loss": 0.2282, | |
"step": 3220 | |
}, | |
{ | |
"epoch": 1.72, | |
"grad_norm": 19.42692756652832, | |
"learning_rate": 2.128888888888889e-05, | |
"loss": 0.2567, | |
"step": 3230 | |
}, | |
{ | |
"epoch": 1.73, | |
"grad_norm": 22.812788009643555, | |
"learning_rate": 2.12e-05, | |
"loss": 0.2344, | |
"step": 3240 | |
}, | |
{ | |
"epoch": 1.73, | |
"grad_norm": 31.932159423828125, | |
"learning_rate": 2.111111111111111e-05, | |
"loss": 0.3359, | |
"step": 3250 | |
}, | |
{ | |
"epoch": 1.74, | |
"grad_norm": 21.950088500976562, | |
"learning_rate": 2.1022222222222225e-05, | |
"loss": 0.2967, | |
"step": 3260 | |
}, | |
{ | |
"epoch": 1.74, | |
"grad_norm": 31.876483917236328, | |
"learning_rate": 2.0933333333333335e-05, | |
"loss": 0.3029, | |
"step": 3270 | |
}, | |
{ | |
"epoch": 1.75, | |
"grad_norm": 28.16164779663086, | |
"learning_rate": 2.0844444444444446e-05, | |
"loss": 0.3843, | |
"step": 3280 | |
}, | |
{ | |
"epoch": 1.75, | |
"grad_norm": 18.941884994506836, | |
"learning_rate": 2.0755555555555556e-05, | |
"loss": 0.3108, | |
"step": 3290 | |
}, | |
{ | |
"epoch": 1.76, | |
"grad_norm": 15.317469596862793, | |
"learning_rate": 2.0666666666666666e-05, | |
"loss": 0.3383, | |
"step": 3300 | |
}, | |
{ | |
"epoch": 1.76, | |
"eval_accuracy": 0.8553, | |
"eval_f1_macro": 0.7873309798668017, | |
"eval_f1_micro": 0.8553, | |
"eval_loss": 0.5249765515327454, | |
"eval_runtime": 125.2578, | |
"eval_samples_per_second": 79.835, | |
"eval_steps_per_second": 2.499, | |
"step": 3300 | |
}, | |
{ | |
"epoch": 1.77, | |
"grad_norm": 21.06130027770996, | |
"learning_rate": 2.057777777777778e-05, | |
"loss": 0.2807, | |
"step": 3310 | |
}, | |
{ | |
"epoch": 1.77, | |
"grad_norm": 24.9218692779541, | |
"learning_rate": 2.048888888888889e-05, | |
"loss": 0.2634, | |
"step": 3320 | |
}, | |
{ | |
"epoch": 1.78, | |
"grad_norm": 16.426345825195312, | |
"learning_rate": 2.04e-05, | |
"loss": 0.2336, | |
"step": 3330 | |
}, | |
{ | |
"epoch": 1.78, | |
"grad_norm": 26.309751510620117, | |
"learning_rate": 2.031111111111111e-05, | |
"loss": 0.3588, | |
"step": 3340 | |
}, | |
{ | |
"epoch": 1.79, | |
"grad_norm": 16.83870506286621, | |
"learning_rate": 2.0222222222222222e-05, | |
"loss": 0.3155, | |
"step": 3350 | |
}, | |
{ | |
"epoch": 1.79, | |
"grad_norm": 19.22264862060547, | |
"learning_rate": 2.0133333333333336e-05, | |
"loss": 0.2739, | |
"step": 3360 | |
}, | |
{ | |
"epoch": 1.8, | |
"grad_norm": 19.02859878540039, | |
"learning_rate": 2.0044444444444446e-05, | |
"loss": 0.2537, | |
"step": 3370 | |
}, | |
{ | |
"epoch": 1.8, | |
"grad_norm": 23.24178695678711, | |
"learning_rate": 1.9955555555555557e-05, | |
"loss": 0.2639, | |
"step": 3380 | |
}, | |
{ | |
"epoch": 1.81, | |
"grad_norm": 13.695367813110352, | |
"learning_rate": 1.9866666666666667e-05, | |
"loss": 0.2067, | |
"step": 3390 | |
}, | |
{ | |
"epoch": 1.81, | |
"grad_norm": 20.969131469726562, | |
"learning_rate": 1.9777777777777778e-05, | |
"loss": 0.1886, | |
"step": 3400 | |
}, | |
{ | |
"epoch": 1.81, | |
"eval_accuracy": 0.8658, | |
"eval_f1_macro": 0.7923625631474337, | |
"eval_f1_micro": 0.8658, | |
"eval_loss": 0.5301083922386169, | |
"eval_runtime": 125.2412, | |
"eval_samples_per_second": 79.846, | |
"eval_steps_per_second": 2.499, | |
"step": 3400 | |
}, | |
{ | |
"epoch": 1.82, | |
"grad_norm": 30.716920852661133, | |
"learning_rate": 1.968888888888889e-05, | |
"loss": 0.3543, | |
"step": 3410 | |
}, | |
{ | |
"epoch": 1.82, | |
"grad_norm": 25.40995216369629, | |
"learning_rate": 1.9600000000000002e-05, | |
"loss": 0.3573, | |
"step": 3420 | |
}, | |
{ | |
"epoch": 1.83, | |
"grad_norm": 25.003276824951172, | |
"learning_rate": 1.9511111111111113e-05, | |
"loss": 0.2405, | |
"step": 3430 | |
}, | |
{ | |
"epoch": 1.83, | |
"grad_norm": 37.33772277832031, | |
"learning_rate": 1.9422222222222223e-05, | |
"loss": 0.3001, | |
"step": 3440 | |
}, | |
{ | |
"epoch": 1.84, | |
"grad_norm": 29.57843589782715, | |
"learning_rate": 1.9333333333333333e-05, | |
"loss": 0.2823, | |
"step": 3450 | |
}, | |
{ | |
"epoch": 1.85, | |
"grad_norm": 23.851917266845703, | |
"learning_rate": 1.9244444444444444e-05, | |
"loss": 0.235, | |
"step": 3460 | |
}, | |
{ | |
"epoch": 1.85, | |
"grad_norm": 25.79085922241211, | |
"learning_rate": 1.9155555555555558e-05, | |
"loss": 0.2136, | |
"step": 3470 | |
}, | |
{ | |
"epoch": 1.86, | |
"grad_norm": 18.118497848510742, | |
"learning_rate": 1.9066666666666668e-05, | |
"loss": 0.2425, | |
"step": 3480 | |
}, | |
{ | |
"epoch": 1.86, | |
"grad_norm": 31.972190856933594, | |
"learning_rate": 1.897777777777778e-05, | |
"loss": 0.2761, | |
"step": 3490 | |
}, | |
{ | |
"epoch": 1.87, | |
"grad_norm": 22.953575134277344, | |
"learning_rate": 1.888888888888889e-05, | |
"loss": 0.2468, | |
"step": 3500 | |
}, | |
{ | |
"epoch": 1.87, | |
"eval_accuracy": 0.8595, | |
"eval_f1_macro": 0.7812713849689391, | |
"eval_f1_micro": 0.8595, | |
"eval_loss": 0.5459020733833313, | |
"eval_runtime": 125.3137, | |
"eval_samples_per_second": 79.8, | |
"eval_steps_per_second": 2.498, | |
"step": 3500 | |
}, | |
{ | |
"epoch": 1.87, | |
"grad_norm": 16.223791122436523, | |
"learning_rate": 1.88e-05, | |
"loss": 0.2574, | |
"step": 3510 | |
}, | |
{ | |
"epoch": 1.88, | |
"grad_norm": 11.477206230163574, | |
"learning_rate": 1.8711111111111113e-05, | |
"loss": 0.1905, | |
"step": 3520 | |
}, | |
{ | |
"epoch": 1.88, | |
"grad_norm": 22.725454330444336, | |
"learning_rate": 1.8622222222222224e-05, | |
"loss": 0.2126, | |
"step": 3530 | |
}, | |
{ | |
"epoch": 1.89, | |
"grad_norm": 13.858516693115234, | |
"learning_rate": 1.8533333333333334e-05, | |
"loss": 0.2417, | |
"step": 3540 | |
}, | |
{ | |
"epoch": 1.89, | |
"grad_norm": 14.678023338317871, | |
"learning_rate": 1.8444444444444445e-05, | |
"loss": 0.2442, | |
"step": 3550 | |
}, | |
{ | |
"epoch": 1.9, | |
"grad_norm": 11.627692222595215, | |
"learning_rate": 1.8355555555555555e-05, | |
"loss": 0.2736, | |
"step": 3560 | |
}, | |
{ | |
"epoch": 1.9, | |
"grad_norm": 24.857446670532227, | |
"learning_rate": 1.826666666666667e-05, | |
"loss": 0.3312, | |
"step": 3570 | |
}, | |
{ | |
"epoch": 1.91, | |
"grad_norm": 22.696962356567383, | |
"learning_rate": 1.817777777777778e-05, | |
"loss": 0.2294, | |
"step": 3580 | |
}, | |
{ | |
"epoch": 1.91, | |
"grad_norm": 33.160343170166016, | |
"learning_rate": 1.808888888888889e-05, | |
"loss": 0.2481, | |
"step": 3590 | |
}, | |
{ | |
"epoch": 1.92, | |
"grad_norm": 21.321407318115234, | |
"learning_rate": 1.8e-05, | |
"loss": 0.2947, | |
"step": 3600 | |
}, | |
{ | |
"epoch": 1.92, | |
"eval_accuracy": 0.8688, | |
"eval_f1_macro": 0.7910490707304132, | |
"eval_f1_micro": 0.8688, | |
"eval_loss": 0.51407390832901, | |
"eval_runtime": 125.3442, | |
"eval_samples_per_second": 79.78, | |
"eval_steps_per_second": 2.497, | |
"step": 3600 | |
}, | |
{ | |
"epoch": 1.93, | |
"grad_norm": 24.3467960357666, | |
"learning_rate": 1.791111111111111e-05, | |
"loss": 0.2795, | |
"step": 3610 | |
}, | |
{ | |
"epoch": 1.93, | |
"grad_norm": 24.753925323486328, | |
"learning_rate": 1.7822222222222225e-05, | |
"loss": 0.2395, | |
"step": 3620 | |
}, | |
{ | |
"epoch": 1.94, | |
"grad_norm": 17.409852981567383, | |
"learning_rate": 1.7733333333333335e-05, | |
"loss": 0.2617, | |
"step": 3630 | |
}, | |
{ | |
"epoch": 1.94, | |
"grad_norm": 24.267776489257812, | |
"learning_rate": 1.7644444444444446e-05, | |
"loss": 0.2099, | |
"step": 3640 | |
}, | |
{ | |
"epoch": 1.95, | |
"grad_norm": 24.462526321411133, | |
"learning_rate": 1.7555555555555556e-05, | |
"loss": 0.2953, | |
"step": 3650 | |
}, | |
{ | |
"epoch": 1.95, | |
"grad_norm": 16.87688446044922, | |
"learning_rate": 1.7466666666666667e-05, | |
"loss": 0.364, | |
"step": 3660 | |
}, | |
{ | |
"epoch": 1.96, | |
"grad_norm": 16.433847427368164, | |
"learning_rate": 1.737777777777778e-05, | |
"loss": 0.2613, | |
"step": 3670 | |
}, | |
{ | |
"epoch": 1.96, | |
"grad_norm": 29.817991256713867, | |
"learning_rate": 1.728888888888889e-05, | |
"loss": 0.3216, | |
"step": 3680 | |
}, | |
{ | |
"epoch": 1.97, | |
"grad_norm": 26.87920379638672, | |
"learning_rate": 1.7199999999999998e-05, | |
"loss": 0.3593, | |
"step": 3690 | |
}, | |
{ | |
"epoch": 1.97, | |
"grad_norm": 20.05802345275879, | |
"learning_rate": 1.7111111111111112e-05, | |
"loss": 0.2625, | |
"step": 3700 | |
}, | |
{ | |
"epoch": 1.97, | |
"eval_accuracy": 0.867, | |
"eval_f1_macro": 0.7909960365885448, | |
"eval_f1_micro": 0.867, | |
"eval_loss": 0.5025032162666321, | |
"eval_runtime": 125.4107, | |
"eval_samples_per_second": 79.738, | |
"eval_steps_per_second": 2.496, | |
"step": 3700 | |
} | |
], | |
"logging_steps": 10, | |
"max_steps": 5625, | |
"num_input_tokens_seen": 0, | |
"num_train_epochs": 3, | |
"save_steps": 100, | |
"total_flos": 6.180866444020941e+17, | |
"train_batch_size": 16, | |
"trial_name": null, | |
"trial_params": null | |
} | |