| { | |
| "best_metric": 0.49125248193740845, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-2340", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 2.631645917892456, | |
| "learning_rate": 1.9923076923076926e-05, | |
| "loss": 1.0806, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 2.7555034160614014, | |
| "learning_rate": 1.9846153846153847e-05, | |
| "loss": 1.0886, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 3.3691883087158203, | |
| "learning_rate": 1.976923076923077e-05, | |
| "loss": 1.0765, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 3.109200954437256, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0931, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 2.8181138038635254, | |
| "learning_rate": 1.9615384615384617e-05, | |
| "loss": 1.0809, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 2.4915010929107666, | |
| "learning_rate": 1.953846153846154e-05, | |
| "loss": 1.0718, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 3.8907675743103027, | |
| "learning_rate": 1.9461538461538462e-05, | |
| "loss": 1.0905, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 3.9334473609924316, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 1.09, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 2.9074010848999023, | |
| "learning_rate": 1.930769230769231e-05, | |
| "loss": 1.0835, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 3.810696840286255, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 1.0851, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 2.3626232147216797, | |
| "learning_rate": 1.9153846153846156e-05, | |
| "loss": 1.0751, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.9881410598754883, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 1.0728, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 8.272817611694336, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.07, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.41353383458646614, | |
| "eval_loss": 1.0682790279388428, | |
| "eval_runtime": 1.1629, | |
| "eval_samples_per_second": 114.366, | |
| "eval_steps_per_second": 14.618, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 2.923727035522461, | |
| "learning_rate": 1.8923076923076925e-05, | |
| "loss": 1.0742, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 4.396390914916992, | |
| "learning_rate": 1.8846153846153846e-05, | |
| "loss": 1.0647, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 2.367124080657959, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 1.0739, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 3.049018621444702, | |
| "learning_rate": 1.8692307692307695e-05, | |
| "loss": 1.0518, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 3.515366554260254, | |
| "learning_rate": 1.8615384615384616e-05, | |
| "loss": 1.0581, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 3.3465802669525146, | |
| "learning_rate": 1.853846153846154e-05, | |
| "loss": 1.0578, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 3.190157175064087, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 1.0674, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 4.284377574920654, | |
| "learning_rate": 1.8384615384615386e-05, | |
| "loss": 1.0656, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 4.362184524536133, | |
| "learning_rate": 1.830769230769231e-05, | |
| "loss": 1.0597, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 3.011751651763916, | |
| "learning_rate": 1.823076923076923e-05, | |
| "loss": 1.05, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 2.408613920211792, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 1.0617, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 16.68673324584961, | |
| "learning_rate": 1.807692307692308e-05, | |
| "loss": 1.0474, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.727813243865967, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.0523, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6240601503759399, | |
| "eval_loss": 1.035576581954956, | |
| "eval_runtime": 0.9256, | |
| "eval_samples_per_second": 143.683, | |
| "eval_steps_per_second": 18.365, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 2.9416935443878174, | |
| "learning_rate": 1.7923076923076925e-05, | |
| "loss": 1.0505, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 3.15810489654541, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 1.0492, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 3.2049734592437744, | |
| "learning_rate": 1.776923076923077e-05, | |
| "loss": 1.0467, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 3.3393309116363525, | |
| "learning_rate": 1.7692307692307694e-05, | |
| "loss": 1.0297, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 2.938563346862793, | |
| "learning_rate": 1.7615384615384615e-05, | |
| "loss": 1.0442, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 2.6086559295654297, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 1.0506, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 3.588426113128662, | |
| "learning_rate": 1.7461538461538464e-05, | |
| "loss": 1.0351, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 2.821547269821167, | |
| "learning_rate": 1.7384615384615385e-05, | |
| "loss": 1.0213, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 3.074611186981201, | |
| "learning_rate": 1.730769230769231e-05, | |
| "loss": 1.0406, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 2.6791045665740967, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 1.0237, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 2.8163557052612305, | |
| "learning_rate": 1.7153846153846155e-05, | |
| "loss": 1.0337, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 3.1033220291137695, | |
| "learning_rate": 1.707692307692308e-05, | |
| "loss": 1.0288, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 6.048186302185059, | |
| "learning_rate": 1.7e-05, | |
| "loss": 1.0439, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6616541353383458, | |
| "eval_loss": 1.0045464038848877, | |
| "eval_runtime": 0.9488, | |
| "eval_samples_per_second": 140.174, | |
| "eval_steps_per_second": 17.917, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 3.378181219100952, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 1.0363, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 3.51971697807312, | |
| "learning_rate": 1.684615384615385e-05, | |
| "loss": 1.0182, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 3.365443706512451, | |
| "learning_rate": 1.676923076923077e-05, | |
| "loss": 1.0095, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 3.7726383209228516, | |
| "learning_rate": 1.6692307692307694e-05, | |
| "loss": 1.005, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 4.8340864181518555, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 1.0283, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 3.5189855098724365, | |
| "learning_rate": 1.653846153846154e-05, | |
| "loss": 1.0159, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 3.1230361461639404, | |
| "learning_rate": 1.6461538461538463e-05, | |
| "loss": 1.0098, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 2.575528860092163, | |
| "learning_rate": 1.6384615384615384e-05, | |
| "loss": 1.0065, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 3.465209722518921, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 1.0066, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 2.744354486465454, | |
| "learning_rate": 1.6230769230769233e-05, | |
| "loss": 0.9909, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 2.649618625640869, | |
| "learning_rate": 1.6153846153846154e-05, | |
| "loss": 1.0106, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 3.7920305728912354, | |
| "learning_rate": 1.607692307692308e-05, | |
| "loss": 0.9962, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 6.38316535949707, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.0056, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_loss": 0.9671366810798645, | |
| "eval_runtime": 0.9722, | |
| "eval_samples_per_second": 136.81, | |
| "eval_steps_per_second": 17.487, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 4.276334285736084, | |
| "learning_rate": 1.5923076923076924e-05, | |
| "loss": 0.9717, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 3.4239840507507324, | |
| "learning_rate": 1.5846153846153848e-05, | |
| "loss": 1.0061, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 3.4140846729278564, | |
| "learning_rate": 1.576923076923077e-05, | |
| "loss": 1.0057, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 3.539780855178833, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 1.0197, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 4.03983211517334, | |
| "learning_rate": 1.5615384615384618e-05, | |
| "loss": 0.9655, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 3.4873383045196533, | |
| "learning_rate": 1.553846153846154e-05, | |
| "loss": 0.9859, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 3.594426393508911, | |
| "learning_rate": 1.5461538461538463e-05, | |
| "loss": 0.9938, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 3.7454657554626465, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.97, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 4.500467300415039, | |
| "learning_rate": 1.5307692307692308e-05, | |
| "loss": 0.976, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 2.860618829727173, | |
| "learning_rate": 1.523076923076923e-05, | |
| "loss": 1.0125, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 4.1896796226501465, | |
| "learning_rate": 1.5153846153846155e-05, | |
| "loss": 0.9496, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 3.3220174312591553, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 0.9541, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 7.564222812652588, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.9853, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7894736842105263, | |
| "eval_loss": 0.9245139956474304, | |
| "eval_runtime": 0.9408, | |
| "eval_samples_per_second": 141.37, | |
| "eval_steps_per_second": 18.07, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.076923076923077, | |
| "grad_norm": 3.3493552207946777, | |
| "learning_rate": 1.4923076923076925e-05, | |
| "loss": 0.9562, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.153846153846154, | |
| "grad_norm": 4.18095588684082, | |
| "learning_rate": 1.4846153846153847e-05, | |
| "loss": 0.9853, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.230769230769231, | |
| "grad_norm": 3.0175259113311768, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.9418, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.3076923076923075, | |
| "grad_norm": 4.064915180206299, | |
| "learning_rate": 1.4692307692307694e-05, | |
| "loss": 0.9644, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 4.862594127655029, | |
| "learning_rate": 1.4615384615384615e-05, | |
| "loss": 0.9521, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.461538461538462, | |
| "grad_norm": 3.188516855239868, | |
| "learning_rate": 1.453846153846154e-05, | |
| "loss": 0.9612, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 5.538461538461538, | |
| "grad_norm": 3.2765607833862305, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.9351, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.615384615384615, | |
| "grad_norm": 2.763471841812134, | |
| "learning_rate": 1.4384615384615387e-05, | |
| "loss": 0.956, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.6923076923076925, | |
| "grad_norm": 4.148519515991211, | |
| "learning_rate": 1.430769230769231e-05, | |
| "loss": 0.9196, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.769230769230769, | |
| "grad_norm": 3.0915586948394775, | |
| "learning_rate": 1.4230769230769232e-05, | |
| "loss": 0.9266, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.846153846153846, | |
| "grad_norm": 3.6650705337524414, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.9385, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.923076923076923, | |
| "grad_norm": 4.015691757202148, | |
| "learning_rate": 1.4076923076923079e-05, | |
| "loss": 0.9356, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 11.575014114379883, | |
| "learning_rate": 1.4e-05, | |
| "loss": 0.9581, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7819548872180451, | |
| "eval_loss": 0.8743670582771301, | |
| "eval_runtime": 0.9201, | |
| "eval_samples_per_second": 144.551, | |
| "eval_steps_per_second": 18.477, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.076923076923077, | |
| "grad_norm": 3.6309585571289062, | |
| "learning_rate": 1.3923076923076924e-05, | |
| "loss": 0.9073, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 4.175745010375977, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.9238, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.230769230769231, | |
| "grad_norm": 3.9674503803253174, | |
| "learning_rate": 1.3769230769230771e-05, | |
| "loss": 0.9014, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.3076923076923075, | |
| "grad_norm": 3.084416627883911, | |
| "learning_rate": 1.3692307692307694e-05, | |
| "loss": 0.9023, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 6.384615384615385, | |
| "grad_norm": 4.2650580406188965, | |
| "learning_rate": 1.3615384615384616e-05, | |
| "loss": 0.8993, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 6.461538461538462, | |
| "grad_norm": 4.604613304138184, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.8902, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 6.538461538461538, | |
| "grad_norm": 4.572556018829346, | |
| "learning_rate": 1.3461538461538463e-05, | |
| "loss": 0.9039, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 6.615384615384615, | |
| "grad_norm": 4.689060211181641, | |
| "learning_rate": 1.3384615384615384e-05, | |
| "loss": 0.8743, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.6923076923076925, | |
| "grad_norm": 4.0536394119262695, | |
| "learning_rate": 1.3307692307692309e-05, | |
| "loss": 0.9188, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.769230769230769, | |
| "grad_norm": 3.3945388793945312, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.9175, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.846153846153846, | |
| "grad_norm": 4.679390907287598, | |
| "learning_rate": 1.3153846153846156e-05, | |
| "loss": 0.8946, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 3.4296305179595947, | |
| "learning_rate": 1.3076923076923078e-05, | |
| "loss": 0.8895, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 6.18251895904541, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.9044, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7819548872180451, | |
| "eval_loss": 0.8171929717063904, | |
| "eval_runtime": 0.9448, | |
| "eval_samples_per_second": 140.765, | |
| "eval_steps_per_second": 17.993, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.076923076923077, | |
| "grad_norm": 4.493593692779541, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.9023, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 7.153846153846154, | |
| "grad_norm": 3.8425426483154297, | |
| "learning_rate": 1.2846153846153848e-05, | |
| "loss": 0.8691, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 7.230769230769231, | |
| "grad_norm": 4.417716026306152, | |
| "learning_rate": 1.2769230769230769e-05, | |
| "loss": 0.8817, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 7.3076923076923075, | |
| "grad_norm": 4.091881275177002, | |
| "learning_rate": 1.2692307692307693e-05, | |
| "loss": 0.8901, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 7.384615384615385, | |
| "grad_norm": 4.237718105316162, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.863, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 7.461538461538462, | |
| "grad_norm": 6.6138410568237305, | |
| "learning_rate": 1.253846153846154e-05, | |
| "loss": 0.8556, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 7.538461538461538, | |
| "grad_norm": 4.715096473693848, | |
| "learning_rate": 1.2461538461538463e-05, | |
| "loss": 0.8851, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 7.615384615384615, | |
| "grad_norm": 3.7215723991394043, | |
| "learning_rate": 1.2384615384615385e-05, | |
| "loss": 0.8334, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 3.3076252937316895, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.8542, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.769230769230769, | |
| "grad_norm": 4.20552921295166, | |
| "learning_rate": 1.2230769230769232e-05, | |
| "loss": 0.8271, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.846153846153846, | |
| "grad_norm": 4.651803016662598, | |
| "learning_rate": 1.2153846153846153e-05, | |
| "loss": 0.8346, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.923076923076923, | |
| "grad_norm": 4.573400974273682, | |
| "learning_rate": 1.2076923076923078e-05, | |
| "loss": 0.8315, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 6.5735182762146, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.869, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_loss": 0.773723304271698, | |
| "eval_runtime": 0.9363, | |
| "eval_samples_per_second": 142.041, | |
| "eval_steps_per_second": 18.156, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.076923076923077, | |
| "grad_norm": 4.215645790100098, | |
| "learning_rate": 1.1923076923076925e-05, | |
| "loss": 0.8822, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.153846153846153, | |
| "grad_norm": 3.5975663661956787, | |
| "learning_rate": 1.1846153846153847e-05, | |
| "loss": 0.8305, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 8.23076923076923, | |
| "grad_norm": 4.703883647918701, | |
| "learning_rate": 1.176923076923077e-05, | |
| "loss": 0.8559, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 8.307692307692308, | |
| "grad_norm": 3.2407641410827637, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.8117, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.384615384615385, | |
| "grad_norm": 4.487489700317383, | |
| "learning_rate": 1.1615384615384617e-05, | |
| "loss": 0.8241, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 3.8978068828582764, | |
| "learning_rate": 1.1538461538461538e-05, | |
| "loss": 0.8537, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.538461538461538, | |
| "grad_norm": 2.5178143978118896, | |
| "learning_rate": 1.1461538461538462e-05, | |
| "loss": 0.8062, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 8.615384615384615, | |
| "grad_norm": 4.473972320556641, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.8532, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 8.692307692307692, | |
| "grad_norm": 5.469933032989502, | |
| "learning_rate": 1.1307692307692309e-05, | |
| "loss": 0.8532, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 8.76923076923077, | |
| "grad_norm": 2.7200920581817627, | |
| "learning_rate": 1.1230769230769232e-05, | |
| "loss": 0.8203, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 8.846153846153847, | |
| "grad_norm": 3.509950876235962, | |
| "learning_rate": 1.1153846153846154e-05, | |
| "loss": 0.82, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.923076923076923, | |
| "grad_norm": 3.90140962600708, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.8182, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 16.341764450073242, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.8804, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_loss": 0.7098143100738525, | |
| "eval_runtime": 0.9276, | |
| "eval_samples_per_second": 143.385, | |
| "eval_steps_per_second": 18.327, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.076923076923077, | |
| "grad_norm": 5.392989635467529, | |
| "learning_rate": 1.0923076923076922e-05, | |
| "loss": 0.8143, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 9.153846153846153, | |
| "grad_norm": 3.0304982662200928, | |
| "learning_rate": 1.0846153846153847e-05, | |
| "loss": 0.7586, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 3.1372382640838623, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.7662, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.307692307692308, | |
| "grad_norm": 3.1745128631591797, | |
| "learning_rate": 1.0692307692307694e-05, | |
| "loss": 0.8316, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 9.384615384615385, | |
| "grad_norm": 6.142508029937744, | |
| "learning_rate": 1.0615384615384616e-05, | |
| "loss": 0.799, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 9.461538461538462, | |
| "grad_norm": 4.660125255584717, | |
| "learning_rate": 1.0538461538461539e-05, | |
| "loss": 0.7981, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 9.538461538461538, | |
| "grad_norm": 3.9741530418395996, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.7887, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "grad_norm": 4.778607368469238, | |
| "learning_rate": 1.0384615384615386e-05, | |
| "loss": 0.7304, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 9.692307692307692, | |
| "grad_norm": 4.286049842834473, | |
| "learning_rate": 1.0307692307692307e-05, | |
| "loss": 0.8002, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 9.76923076923077, | |
| "grad_norm": 3.7198238372802734, | |
| "learning_rate": 1.0230769230769231e-05, | |
| "loss": 0.7861, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 9.846153846153847, | |
| "grad_norm": 3.3311705589294434, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.7667, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 9.923076923076923, | |
| "grad_norm": 4.488588333129883, | |
| "learning_rate": 1.0076923076923078e-05, | |
| "loss": 0.8, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 8.365084648132324, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7757, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8120300751879699, | |
| "eval_loss": 0.6705044507980347, | |
| "eval_runtime": 0.9309, | |
| "eval_samples_per_second": 142.866, | |
| "eval_steps_per_second": 18.261, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.076923076923077, | |
| "grad_norm": 4.813228130340576, | |
| "learning_rate": 9.923076923076923e-06, | |
| "loss": 0.7808, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 10.153846153846153, | |
| "grad_norm": 3.86871337890625, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.7749, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 10.23076923076923, | |
| "grad_norm": 4.498291492462158, | |
| "learning_rate": 9.76923076923077e-06, | |
| "loss": 0.7842, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 10.307692307692308, | |
| "grad_norm": 7.416966438293457, | |
| "learning_rate": 9.692307692307693e-06, | |
| "loss": 0.7397, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 10.384615384615385, | |
| "grad_norm": 3.8194034099578857, | |
| "learning_rate": 9.615384615384616e-06, | |
| "loss": 0.7101, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 10.461538461538462, | |
| "grad_norm": 4.42163610458374, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.7694, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 10.538461538461538, | |
| "grad_norm": 4.710392951965332, | |
| "learning_rate": 9.461538461538463e-06, | |
| "loss": 0.7536, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 10.615384615384615, | |
| "grad_norm": 4.0583906173706055, | |
| "learning_rate": 9.384615384615385e-06, | |
| "loss": 0.7591, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 10.692307692307692, | |
| "grad_norm": 4.696585655212402, | |
| "learning_rate": 9.307692307692308e-06, | |
| "loss": 0.7469, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "grad_norm": 4.995838642120361, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.776, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.846153846153847, | |
| "grad_norm": 3.7382941246032715, | |
| "learning_rate": 9.153846153846155e-06, | |
| "loss": 0.7281, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 10.923076923076923, | |
| "grad_norm": 4.595252513885498, | |
| "learning_rate": 9.076923076923078e-06, | |
| "loss": 0.7781, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 10.765867233276367, | |
| "learning_rate": 9e-06, | |
| "loss": 0.7694, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.638173520565033, | |
| "eval_runtime": 0.9524, | |
| "eval_samples_per_second": 139.647, | |
| "eval_steps_per_second": 17.85, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 11.076923076923077, | |
| "grad_norm": 4.3198041915893555, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.7603, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 11.153846153846153, | |
| "grad_norm": 6.03615140914917, | |
| "learning_rate": 8.846153846153847e-06, | |
| "loss": 0.7594, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 11.23076923076923, | |
| "grad_norm": 4.512632846832275, | |
| "learning_rate": 8.76923076923077e-06, | |
| "loss": 0.7111, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 11.307692307692308, | |
| "grad_norm": 3.5640311241149902, | |
| "learning_rate": 8.692307692307692e-06, | |
| "loss": 0.7783, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 11.384615384615385, | |
| "grad_norm": 3.612410306930542, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.7285, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 11.461538461538462, | |
| "grad_norm": 5.0590996742248535, | |
| "learning_rate": 8.53846153846154e-06, | |
| "loss": 0.7128, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "grad_norm": 4.820272445678711, | |
| "learning_rate": 8.461538461538462e-06, | |
| "loss": 0.798, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.615384615384615, | |
| "grad_norm": 3.6416468620300293, | |
| "learning_rate": 8.384615384615385e-06, | |
| "loss": 0.7408, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 11.692307692307692, | |
| "grad_norm": 4.562481880187988, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.7682, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 11.76923076923077, | |
| "grad_norm": 3.1171770095825195, | |
| "learning_rate": 8.230769230769232e-06, | |
| "loss": 0.7337, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 11.846153846153847, | |
| "grad_norm": 3.2231085300445557, | |
| "learning_rate": 8.153846153846154e-06, | |
| "loss": 0.7348, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 11.923076923076923, | |
| "grad_norm": 3.5213522911071777, | |
| "learning_rate": 8.076923076923077e-06, | |
| "loss": 0.7064, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 7.278842449188232, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.7966, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7894736842105263, | |
| "eval_loss": 0.6087508797645569, | |
| "eval_runtime": 0.9347, | |
| "eval_samples_per_second": 142.285, | |
| "eval_steps_per_second": 18.187, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 12.076923076923077, | |
| "grad_norm": 8.028270721435547, | |
| "learning_rate": 7.923076923076924e-06, | |
| "loss": 0.7035, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 12.153846153846153, | |
| "grad_norm": 3.4982223510742188, | |
| "learning_rate": 7.846153846153847e-06, | |
| "loss": 0.6953, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 12.23076923076923, | |
| "grad_norm": 3.474436044692993, | |
| "learning_rate": 7.76923076923077e-06, | |
| "loss": 0.7272, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 5.147262096405029, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.7639, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.384615384615385, | |
| "grad_norm": 3.383554458618164, | |
| "learning_rate": 7.615384615384615e-06, | |
| "loss": 0.7094, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 12.461538461538462, | |
| "grad_norm": 3.3019802570343018, | |
| "learning_rate": 7.538461538461539e-06, | |
| "loss": 0.685, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 12.538461538461538, | |
| "grad_norm": 3.602853775024414, | |
| "learning_rate": 7.461538461538462e-06, | |
| "loss": 0.7068, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 12.615384615384615, | |
| "grad_norm": 3.3312697410583496, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.7045, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 12.692307692307692, | |
| "grad_norm": 6.843470573425293, | |
| "learning_rate": 7.307692307692308e-06, | |
| "loss": 0.6643, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 12.76923076923077, | |
| "grad_norm": 9.701897621154785, | |
| "learning_rate": 7.230769230769231e-06, | |
| "loss": 0.8052, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 12.846153846153847, | |
| "grad_norm": 5.923687934875488, | |
| "learning_rate": 7.153846153846155e-06, | |
| "loss": 0.7118, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 12.923076923076923, | |
| "grad_norm": 3.330617666244507, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.701, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 10.613458633422852, | |
| "learning_rate": 7e-06, | |
| "loss": 0.7425, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_loss": 0.572424054145813, | |
| "eval_runtime": 0.9506, | |
| "eval_samples_per_second": 139.913, | |
| "eval_steps_per_second": 17.884, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 13.076923076923077, | |
| "grad_norm": 4.795916557312012, | |
| "learning_rate": 6.923076923076923e-06, | |
| "loss": 0.7275, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.153846153846153, | |
| "grad_norm": 3.3990397453308105, | |
| "learning_rate": 6.846153846153847e-06, | |
| "loss": 0.737, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 13.23076923076923, | |
| "grad_norm": 5.3640851974487305, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.7653, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 13.307692307692308, | |
| "grad_norm": 2.9321601390838623, | |
| "learning_rate": 6.692307692307692e-06, | |
| "loss": 0.741, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 13.384615384615385, | |
| "grad_norm": 5.113746166229248, | |
| "learning_rate": 6.615384615384616e-06, | |
| "loss": 0.6832, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 13.461538461538462, | |
| "grad_norm": 4.589268207550049, | |
| "learning_rate": 6.538461538461539e-06, | |
| "loss": 0.7463, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 13.538461538461538, | |
| "grad_norm": 3.7653889656066895, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.7569, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 13.615384615384615, | |
| "grad_norm": 4.248018264770508, | |
| "learning_rate": 6.384615384615384e-06, | |
| "loss": 0.6853, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 13.692307692307692, | |
| "grad_norm": 4.481900691986084, | |
| "learning_rate": 6.307692307692308e-06, | |
| "loss": 0.6679, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 13.76923076923077, | |
| "grad_norm": 3.7759058475494385, | |
| "learning_rate": 6.230769230769231e-06, | |
| "loss": 0.7159, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "grad_norm": 7.013620853424072, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.6663, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.923076923076923, | |
| "grad_norm": 3.7434396743774414, | |
| "learning_rate": 6.076923076923077e-06, | |
| "loss": 0.6043, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 8.63290023803711, | |
| "learning_rate": 6e-06, | |
| "loss": 0.7698, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_loss": 0.5665194392204285, | |
| "eval_runtime": 0.9506, | |
| "eval_samples_per_second": 139.918, | |
| "eval_steps_per_second": 17.884, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 14.076923076923077, | |
| "grad_norm": 6.073098659515381, | |
| "learning_rate": 5.923076923076924e-06, | |
| "loss": 0.701, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 14.153846153846153, | |
| "grad_norm": 4.466890811920166, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.712, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 14.23076923076923, | |
| "grad_norm": 4.844086170196533, | |
| "learning_rate": 5.769230769230769e-06, | |
| "loss": 0.6637, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 14.307692307692308, | |
| "grad_norm": 5.5982184410095215, | |
| "learning_rate": 5.692307692307692e-06, | |
| "loss": 0.6735, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 14.384615384615385, | |
| "grad_norm": 6.027699947357178, | |
| "learning_rate": 5.615384615384616e-06, | |
| "loss": 0.672, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 14.461538461538462, | |
| "grad_norm": 3.128363847732544, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.6864, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 14.538461538461538, | |
| "grad_norm": 5.279370307922363, | |
| "learning_rate": 5.461538461538461e-06, | |
| "loss": 0.6765, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 14.615384615384615, | |
| "grad_norm": 3.779651165008545, | |
| "learning_rate": 5.384615384615385e-06, | |
| "loss": 0.6902, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 14.692307692307692, | |
| "grad_norm": 7.5796003341674805, | |
| "learning_rate": 5.307692307692308e-06, | |
| "loss": 0.6971, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 14.76923076923077, | |
| "grad_norm": 5.17158842086792, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.6997, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 14.846153846153847, | |
| "grad_norm": 3.0330209732055664, | |
| "learning_rate": 5.1538461538461534e-06, | |
| "loss": 0.778, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 14.923076923076923, | |
| "grad_norm": 7.733983993530273, | |
| "learning_rate": 5.076923076923077e-06, | |
| "loss": 0.6478, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 7.919536590576172, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6632, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.5307806730270386, | |
| "eval_runtime": 0.9229, | |
| "eval_samples_per_second": 144.113, | |
| "eval_steps_per_second": 18.421, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 15.076923076923077, | |
| "grad_norm": 4.707977771759033, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.6156, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 15.153846153846153, | |
| "grad_norm": 5.254363536834717, | |
| "learning_rate": 4.8461538461538465e-06, | |
| "loss": 0.6478, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 15.23076923076923, | |
| "grad_norm": 4.477289199829102, | |
| "learning_rate": 4.76923076923077e-06, | |
| "loss": 0.6786, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 15.307692307692308, | |
| "grad_norm": 6.381091117858887, | |
| "learning_rate": 4.692307692307693e-06, | |
| "loss": 0.7183, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 3.8760223388671875, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.6391, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.461538461538462, | |
| "grad_norm": 7.011289119720459, | |
| "learning_rate": 4.538461538461539e-06, | |
| "loss": 0.6135, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 15.538461538461538, | |
| "grad_norm": 5.04637336730957, | |
| "learning_rate": 4.461538461538462e-06, | |
| "loss": 0.628, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 15.615384615384615, | |
| "grad_norm": 7.818150520324707, | |
| "learning_rate": 4.384615384615385e-06, | |
| "loss": 0.6537, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 15.692307692307692, | |
| "grad_norm": 3.9413349628448486, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.6662, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 15.76923076923077, | |
| "grad_norm": 5.911675930023193, | |
| "learning_rate": 4.230769230769231e-06, | |
| "loss": 0.6284, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 15.846153846153847, | |
| "grad_norm": 4.451909065246582, | |
| "learning_rate": 4.1538461538461545e-06, | |
| "loss": 0.6906, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 15.923076923076923, | |
| "grad_norm": 5.271285057067871, | |
| "learning_rate": 4.076923076923077e-06, | |
| "loss": 0.6453, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 8.229835510253906, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.6162, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_loss": 0.5261984467506409, | |
| "eval_runtime": 0.9205, | |
| "eval_samples_per_second": 144.487, | |
| "eval_steps_per_second": 18.468, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 16.076923076923077, | |
| "grad_norm": 12.219758033752441, | |
| "learning_rate": 3.923076923076923e-06, | |
| "loss": 0.7381, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 16.153846153846153, | |
| "grad_norm": 4.024055480957031, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.708, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.23076923076923, | |
| "grad_norm": 5.345643043518066, | |
| "learning_rate": 3.7692307692307694e-06, | |
| "loss": 0.7214, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 16.307692307692307, | |
| "grad_norm": 7.9586992263793945, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.7163, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 16.384615384615383, | |
| "grad_norm": 7.007926940917969, | |
| "learning_rate": 3.6153846153846156e-06, | |
| "loss": 0.6692, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 16.46153846153846, | |
| "grad_norm": 8.54269027709961, | |
| "learning_rate": 3.538461538461539e-06, | |
| "loss": 0.661, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 16.53846153846154, | |
| "grad_norm": 6.417664051055908, | |
| "learning_rate": 3.4615384615384617e-06, | |
| "loss": 0.6186, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 16.615384615384617, | |
| "grad_norm": 3.766784429550171, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.6254, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 16.692307692307693, | |
| "grad_norm": 4.219268798828125, | |
| "learning_rate": 3.307692307692308e-06, | |
| "loss": 0.6308, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 16.76923076923077, | |
| "grad_norm": 6.51754903793335, | |
| "learning_rate": 3.2307692307692313e-06, | |
| "loss": 0.7395, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 16.846153846153847, | |
| "grad_norm": 4.084061622619629, | |
| "learning_rate": 3.153846153846154e-06, | |
| "loss": 0.7093, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "grad_norm": 2.852893590927124, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.6213, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 9.559004783630371, | |
| "learning_rate": 3e-06, | |
| "loss": 0.6128, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_loss": 0.5081294178962708, | |
| "eval_runtime": 0.9208, | |
| "eval_samples_per_second": 144.443, | |
| "eval_steps_per_second": 18.463, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 17.076923076923077, | |
| "grad_norm": 5.92500114440918, | |
| "learning_rate": 2.9230769230769236e-06, | |
| "loss": 0.6286, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 17.153846153846153, | |
| "grad_norm": 5.656437397003174, | |
| "learning_rate": 2.846153846153846e-06, | |
| "loss": 0.6561, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 17.23076923076923, | |
| "grad_norm": 5.492016792297363, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.6089, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 17.307692307692307, | |
| "grad_norm": 5.833240985870361, | |
| "learning_rate": 2.6923076923076923e-06, | |
| "loss": 0.6818, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 17.384615384615383, | |
| "grad_norm": 3.1228320598602295, | |
| "learning_rate": 2.615384615384616e-06, | |
| "loss": 0.6544, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 17.46153846153846, | |
| "grad_norm": 5.364021301269531, | |
| "learning_rate": 2.5384615384615385e-06, | |
| "loss": 0.5678, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 17.53846153846154, | |
| "grad_norm": 5.820409297943115, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.7722, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 17.615384615384617, | |
| "grad_norm": 4.316230297088623, | |
| "learning_rate": 2.384615384615385e-06, | |
| "loss": 0.6169, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 17.692307692307693, | |
| "grad_norm": 4.034291744232178, | |
| "learning_rate": 2.307692307692308e-06, | |
| "loss": 0.616, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 17.76923076923077, | |
| "grad_norm": 9.319421768188477, | |
| "learning_rate": 2.230769230769231e-06, | |
| "loss": 0.6772, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 17.846153846153847, | |
| "grad_norm": 4.886960029602051, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.6425, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 17.923076923076923, | |
| "grad_norm": 3.6336989402770996, | |
| "learning_rate": 2.0769230769230773e-06, | |
| "loss": 0.6684, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 6.341444492340088, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.685, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.49125248193740845, | |
| "eval_runtime": 0.9184, | |
| "eval_samples_per_second": 144.82, | |
| "eval_steps_per_second": 18.511, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 18.076923076923077, | |
| "grad_norm": 4.5043158531188965, | |
| "learning_rate": 1.9230769230769234e-06, | |
| "loss": 0.5884, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 18.153846153846153, | |
| "grad_norm": 5.0124592781066895, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.7235, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 18.23076923076923, | |
| "grad_norm": 3.5005030632019043, | |
| "learning_rate": 1.7692307692307695e-06, | |
| "loss": 0.6181, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 18.307692307692307, | |
| "grad_norm": 5.270286560058594, | |
| "learning_rate": 1.6923076923076926e-06, | |
| "loss": 0.6754, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 18.384615384615383, | |
| "grad_norm": 3.9892470836639404, | |
| "learning_rate": 1.6153846153846157e-06, | |
| "loss": 0.6778, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 18.46153846153846, | |
| "grad_norm": 4.945601940155029, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.6981, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.53846153846154, | |
| "grad_norm": 6.190303802490234, | |
| "learning_rate": 1.4615384615384618e-06, | |
| "loss": 0.6147, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 18.615384615384617, | |
| "grad_norm": 11.984601974487305, | |
| "learning_rate": 1.3846153846153848e-06, | |
| "loss": 0.695, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 18.692307692307693, | |
| "grad_norm": 5.748422622680664, | |
| "learning_rate": 1.307692307692308e-06, | |
| "loss": 0.6945, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 18.76923076923077, | |
| "grad_norm": 7.194711685180664, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.7725, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 18.846153846153847, | |
| "grad_norm": 4.310044288635254, | |
| "learning_rate": 1.153846153846154e-06, | |
| "loss": 0.6847, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 18.923076923076923, | |
| "grad_norm": 4.310553073883057, | |
| "learning_rate": 1.076923076923077e-06, | |
| "loss": 0.7078, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 7.30472993850708, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.6614, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_loss": 0.49367815256118774, | |
| "eval_runtime": 0.9215, | |
| "eval_samples_per_second": 144.328, | |
| "eval_steps_per_second": 18.448, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 19.076923076923077, | |
| "grad_norm": 3.58013653755188, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.621, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 19.153846153846153, | |
| "grad_norm": 5.2532453536987305, | |
| "learning_rate": 8.461538461538463e-07, | |
| "loss": 0.6367, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 19.23076923076923, | |
| "grad_norm": 11.078712463378906, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 0.6835, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 19.307692307692307, | |
| "grad_norm": 4.118062973022461, | |
| "learning_rate": 6.923076923076924e-07, | |
| "loss": 0.6719, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 19.384615384615383, | |
| "grad_norm": 4.630733966827393, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.6546, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 19.46153846153846, | |
| "grad_norm": 4.452598571777344, | |
| "learning_rate": 5.384615384615386e-07, | |
| "loss": 0.6052, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 19.53846153846154, | |
| "grad_norm": 5.847692966461182, | |
| "learning_rate": 4.615384615384616e-07, | |
| "loss": 0.6786, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 19.615384615384617, | |
| "grad_norm": 5.957764625549316, | |
| "learning_rate": 3.846153846153847e-07, | |
| "loss": 0.677, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 19.692307692307693, | |
| "grad_norm": 6.010245323181152, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.6106, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 19.76923076923077, | |
| "grad_norm": 5.3892412185668945, | |
| "learning_rate": 2.307692307692308e-07, | |
| "loss": 0.5836, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 19.846153846153847, | |
| "grad_norm": 4.214204788208008, | |
| "learning_rate": 1.5384615384615387e-07, | |
| "loss": 0.6448, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 19.923076923076923, | |
| "grad_norm": 4.230797290802002, | |
| "learning_rate": 7.692307692307694e-08, | |
| "loss": 0.5617, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 10.618528366088867, | |
| "learning_rate": 0.0, | |
| "loss": 0.6934, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.5026500821113586, | |
| "eval_runtime": 1.0302, | |
| "eval_samples_per_second": 129.097, | |
| "eval_steps_per_second": 16.501, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2600, | |
| "total_flos": 7.939121542823117e+17, | |
| "train_loss": 0.8216404274793772, | |
| "train_runtime": 338.1702, | |
| "train_samples_per_second": 61.153, | |
| "train_steps_per_second": 7.688 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.939121542823117e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |