| { | |
| "best_metric": 0.0014008664293214679, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-1100", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 1160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.034482758620689655, | |
| "grad_norm": 1.4541093111038208, | |
| "learning_rate": 0.00019827586206896554, | |
| "loss": 0.9417, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06896551724137931, | |
| "grad_norm": 0.6665103435516357, | |
| "learning_rate": 0.00019655172413793104, | |
| "loss": 0.3226, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10344827586206896, | |
| "grad_norm": 0.33402901887893677, | |
| "learning_rate": 0.00019482758620689657, | |
| "loss": 0.1528, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13793103448275862, | |
| "grad_norm": 0.21010038256645203, | |
| "learning_rate": 0.0001931034482758621, | |
| "loss": 0.1077, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1724137931034483, | |
| "grad_norm": 0.2470974624156952, | |
| "learning_rate": 0.0001913793103448276, | |
| "loss": 0.0676, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.20689655172413793, | |
| "grad_norm": 0.16962118446826935, | |
| "learning_rate": 0.00018965517241379312, | |
| "loss": 0.0449, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2413793103448276, | |
| "grad_norm": 0.1308588832616806, | |
| "learning_rate": 0.00018793103448275865, | |
| "loss": 0.0346, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "grad_norm": 0.11898835003376007, | |
| "learning_rate": 0.00018620689655172415, | |
| "loss": 0.0318, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3103448275862069, | |
| "grad_norm": 0.10036325454711914, | |
| "learning_rate": 0.00018448275862068968, | |
| "loss": 0.0425, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3448275862068966, | |
| "grad_norm": 0.10104402154684067, | |
| "learning_rate": 0.00018275862068965518, | |
| "loss": 0.0532, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3448275862068966, | |
| "eval_accuracy": 0.9961315280464217, | |
| "eval_loss": 0.027677614241838455, | |
| "eval_runtime": 9.322, | |
| "eval_samples_per_second": 55.46, | |
| "eval_steps_per_second": 6.973, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3793103448275862, | |
| "grad_norm": 1.163355827331543, | |
| "learning_rate": 0.0001810344827586207, | |
| "loss": 0.1062, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 0.08375360816717148, | |
| "learning_rate": 0.0001793103448275862, | |
| "loss": 0.0258, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4482758620689655, | |
| "grad_norm": 0.09952229261398315, | |
| "learning_rate": 0.00017758620689655173, | |
| "loss": 0.0175, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4827586206896552, | |
| "grad_norm": 0.6083604097366333, | |
| "learning_rate": 0.00017586206896551723, | |
| "loss": 0.0683, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5172413793103449, | |
| "grad_norm": 0.09937556833028793, | |
| "learning_rate": 0.00017413793103448276, | |
| "loss": 0.0947, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "grad_norm": 0.08655019849538803, | |
| "learning_rate": 0.00017241379310344826, | |
| "loss": 0.0201, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5862068965517241, | |
| "grad_norm": 0.059486184269189835, | |
| "learning_rate": 0.0001706896551724138, | |
| "loss": 0.0631, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6206896551724138, | |
| "grad_norm": 0.07637613266706467, | |
| "learning_rate": 0.00016896551724137932, | |
| "loss": 0.032, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.6551724137931034, | |
| "grad_norm": 10.693470001220703, | |
| "learning_rate": 0.00016724137931034482, | |
| "loss": 0.0542, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 0.09118778258562088, | |
| "learning_rate": 0.00016551724137931035, | |
| "loss": 0.0518, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "eval_accuracy": 0.9941972920696325, | |
| "eval_loss": 0.03303845226764679, | |
| "eval_runtime": 9.2358, | |
| "eval_samples_per_second": 55.978, | |
| "eval_steps_per_second": 7.038, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7241379310344828, | |
| "grad_norm": 0.04961516335606575, | |
| "learning_rate": 0.00016379310344827587, | |
| "loss": 0.0708, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.7586206896551724, | |
| "grad_norm": 0.058188311755657196, | |
| "learning_rate": 0.00016206896551724137, | |
| "loss": 0.0392, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7931034482758621, | |
| "grad_norm": 0.041994284838438034, | |
| "learning_rate": 0.0001603448275862069, | |
| "loss": 0.0116, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "grad_norm": 0.05082221329212189, | |
| "learning_rate": 0.00015862068965517243, | |
| "loss": 0.0376, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8620689655172413, | |
| "grad_norm": 0.07286480814218521, | |
| "learning_rate": 0.00015689655172413793, | |
| "loss": 0.0382, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.896551724137931, | |
| "grad_norm": 0.03850896656513214, | |
| "learning_rate": 0.00015517241379310346, | |
| "loss": 0.0178, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.9310344827586207, | |
| "grad_norm": 0.031015470623970032, | |
| "learning_rate": 0.00015344827586206899, | |
| "loss": 0.0328, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.9655172413793104, | |
| "grad_norm": 0.03393784165382385, | |
| "learning_rate": 0.00015172413793103449, | |
| "loss": 0.0072, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.09719406068325043, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.0071, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.0344827586206897, | |
| "grad_norm": 0.02531067281961441, | |
| "learning_rate": 0.00014827586206896554, | |
| "loss": 0.006, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0344827586206897, | |
| "eval_accuracy": 0.9922630560928434, | |
| "eval_loss": 0.038734398782253265, | |
| "eval_runtime": 9.1377, | |
| "eval_samples_per_second": 56.579, | |
| "eval_steps_per_second": 7.113, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0689655172413792, | |
| "grad_norm": 0.02636132761836052, | |
| "learning_rate": 0.00014655172413793104, | |
| "loss": 0.0057, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.103448275862069, | |
| "grad_norm": 0.023396922275424004, | |
| "learning_rate": 0.00014482758620689657, | |
| "loss": 0.0053, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.1379310344827587, | |
| "grad_norm": 0.022781934589147568, | |
| "learning_rate": 0.0001431034482758621, | |
| "loss": 0.005, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.1724137931034484, | |
| "grad_norm": 0.020923634991049767, | |
| "learning_rate": 0.0001413793103448276, | |
| "loss": 0.0048, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.206896551724138, | |
| "grad_norm": 0.020241938531398773, | |
| "learning_rate": 0.0001396551724137931, | |
| "loss": 0.0046, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2413793103448276, | |
| "grad_norm": 0.0188324972987175, | |
| "learning_rate": 0.00013793103448275863, | |
| "loss": 0.0046, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.2758620689655173, | |
| "grad_norm": 0.019414547830820084, | |
| "learning_rate": 0.00013620689655172413, | |
| "loss": 0.0043, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.3103448275862069, | |
| "grad_norm": 0.019972946494817734, | |
| "learning_rate": 0.00013448275862068965, | |
| "loss": 0.0075, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.3448275862068966, | |
| "grad_norm": 0.02629067189991474, | |
| "learning_rate": 0.00013275862068965518, | |
| "loss": 0.0377, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "grad_norm": 0.027980973944067955, | |
| "learning_rate": 0.00013103448275862068, | |
| "loss": 0.004, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "eval_accuracy": 0.9845261121856866, | |
| "eval_loss": 0.0695781335234642, | |
| "eval_runtime": 9.2324, | |
| "eval_samples_per_second": 55.998, | |
| "eval_steps_per_second": 7.04, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4137931034482758, | |
| "grad_norm": 10.141643524169922, | |
| "learning_rate": 0.0001293103448275862, | |
| "loss": 0.0441, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.4482758620689655, | |
| "grad_norm": 0.017165783792734146, | |
| "learning_rate": 0.00012758620689655174, | |
| "loss": 0.0049, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.4827586206896552, | |
| "grad_norm": 0.01630399189889431, | |
| "learning_rate": 0.00012586206896551724, | |
| "loss": 0.0492, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.5172413793103448, | |
| "grad_norm": 0.018257714807987213, | |
| "learning_rate": 0.00012413793103448277, | |
| "loss": 0.0526, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.5517241379310345, | |
| "grad_norm": 0.01886160485446453, | |
| "learning_rate": 0.00012241379310344827, | |
| "loss": 0.0402, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5862068965517242, | |
| "grad_norm": 0.01721111685037613, | |
| "learning_rate": 0.0001206896551724138, | |
| "loss": 0.0359, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.6206896551724137, | |
| "grad_norm": 0.020360512658953667, | |
| "learning_rate": 0.00011896551724137932, | |
| "loss": 0.0051, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.6551724137931034, | |
| "grad_norm": 0.01872898079454899, | |
| "learning_rate": 0.00011724137931034482, | |
| "loss": 0.004, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.6896551724137931, | |
| "grad_norm": 0.015705358237028122, | |
| "learning_rate": 0.00011551724137931035, | |
| "loss": 0.0036, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "grad_norm": 0.022712524980306625, | |
| "learning_rate": 0.00011379310344827588, | |
| "loss": 0.0032, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "eval_accuracy": 0.9690522243713733, | |
| "eval_loss": 0.16091904044151306, | |
| "eval_runtime": 9.2258, | |
| "eval_samples_per_second": 56.039, | |
| "eval_steps_per_second": 7.045, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.7586206896551724, | |
| "grad_norm": 0.02154046855866909, | |
| "learning_rate": 0.00011206896551724138, | |
| "loss": 0.0837, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.793103448275862, | |
| "grad_norm": 0.01605401560664177, | |
| "learning_rate": 0.0001103448275862069, | |
| "loss": 0.0033, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.8275862068965516, | |
| "grad_norm": 0.04890372231602669, | |
| "learning_rate": 0.00010862068965517242, | |
| "loss": 0.0431, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.8620689655172413, | |
| "grad_norm": 0.017495185136795044, | |
| "learning_rate": 0.00010689655172413792, | |
| "loss": 0.0038, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.896551724137931, | |
| "grad_norm": 0.017458785325288773, | |
| "learning_rate": 0.00010517241379310345, | |
| "loss": 0.0036, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.9310344827586206, | |
| "grad_norm": 0.01829078048467636, | |
| "learning_rate": 0.00010344827586206898, | |
| "loss": 0.003, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.9655172413793105, | |
| "grad_norm": 0.015696866437792778, | |
| "learning_rate": 0.00010172413793103448, | |
| "loss": 0.0029, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.015601696446537971, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0027, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.0344827586206895, | |
| "grad_norm": 0.014859354123473167, | |
| "learning_rate": 9.827586206896552e-05, | |
| "loss": 0.0027, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.0689655172413794, | |
| "grad_norm": 0.013295358046889305, | |
| "learning_rate": 9.655172413793105e-05, | |
| "loss": 0.0026, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.0689655172413794, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0025033194106072187, | |
| "eval_runtime": 9.5465, | |
| "eval_samples_per_second": 54.156, | |
| "eval_steps_per_second": 6.809, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.103448275862069, | |
| "grad_norm": 0.015505661256611347, | |
| "learning_rate": 9.482758620689656e-05, | |
| "loss": 0.0025, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.1379310344827585, | |
| "grad_norm": 0.010766665451228619, | |
| "learning_rate": 9.310344827586207e-05, | |
| "loss": 0.0024, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.1724137931034484, | |
| "grad_norm": 0.010864505544304848, | |
| "learning_rate": 9.137931034482759e-05, | |
| "loss": 0.0023, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.206896551724138, | |
| "grad_norm": 0.009982941672205925, | |
| "learning_rate": 8.96551724137931e-05, | |
| "loss": 0.0023, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.2413793103448274, | |
| "grad_norm": 0.009656297974288464, | |
| "learning_rate": 8.793103448275862e-05, | |
| "loss": 0.0022, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.2758620689655173, | |
| "grad_norm": 0.009799299761652946, | |
| "learning_rate": 8.620689655172413e-05, | |
| "loss": 0.0022, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.310344827586207, | |
| "grad_norm": 0.01052001304924488, | |
| "learning_rate": 8.448275862068966e-05, | |
| "loss": 0.0022, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.344827586206897, | |
| "grad_norm": 0.009396790526807308, | |
| "learning_rate": 8.275862068965517e-05, | |
| "loss": 0.0021, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.3793103448275863, | |
| "grad_norm": 0.00975924450904131, | |
| "learning_rate": 8.103448275862069e-05, | |
| "loss": 0.0021, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.413793103448276, | |
| "grad_norm": 0.009396582841873169, | |
| "learning_rate": 7.931034482758621e-05, | |
| "loss": 0.0021, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.413793103448276, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.002028640592470765, | |
| "eval_runtime": 9.6653, | |
| "eval_samples_per_second": 53.49, | |
| "eval_steps_per_second": 6.725, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.4482758620689653, | |
| "grad_norm": 0.008813206106424332, | |
| "learning_rate": 7.758620689655173e-05, | |
| "loss": 0.002, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.4827586206896552, | |
| "grad_norm": 0.008702595718204975, | |
| "learning_rate": 7.586206896551724e-05, | |
| "loss": 0.002, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.5172413793103448, | |
| "grad_norm": 0.010061556473374367, | |
| "learning_rate": 7.413793103448277e-05, | |
| "loss": 0.002, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.5517241379310347, | |
| "grad_norm": 0.010726147331297398, | |
| "learning_rate": 7.241379310344828e-05, | |
| "loss": 0.0019, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.586206896551724, | |
| "grad_norm": 0.009536409750580788, | |
| "learning_rate": 7.06896551724138e-05, | |
| "loss": 0.0019, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.6206896551724137, | |
| "grad_norm": 0.008245566859841347, | |
| "learning_rate": 6.896551724137931e-05, | |
| "loss": 0.0019, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.655172413793103, | |
| "grad_norm": 0.010203349404036999, | |
| "learning_rate": 6.724137931034483e-05, | |
| "loss": 0.0019, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.689655172413793, | |
| "grad_norm": 0.008181954734027386, | |
| "learning_rate": 6.551724137931034e-05, | |
| "loss": 0.0018, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.7241379310344827, | |
| "grad_norm": 0.008396759629249573, | |
| "learning_rate": 6.379310344827587e-05, | |
| "loss": 0.0018, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.7586206896551726, | |
| "grad_norm": 0.008574232459068298, | |
| "learning_rate": 6.206896551724138e-05, | |
| "loss": 0.0018, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.7586206896551726, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0017438590293750167, | |
| "eval_runtime": 9.2473, | |
| "eval_samples_per_second": 55.908, | |
| "eval_steps_per_second": 7.029, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.793103448275862, | |
| "grad_norm": 0.008582869544625282, | |
| "learning_rate": 6.03448275862069e-05, | |
| "loss": 0.0017, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.8275862068965516, | |
| "grad_norm": 0.00796087272465229, | |
| "learning_rate": 5.862068965517241e-05, | |
| "loss": 0.0017, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.862068965517241, | |
| "grad_norm": 0.00767522631213069, | |
| "learning_rate": 5.689655172413794e-05, | |
| "loss": 0.0017, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.896551724137931, | |
| "grad_norm": 0.007729759905487299, | |
| "learning_rate": 5.517241379310345e-05, | |
| "loss": 0.0017, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.9310344827586206, | |
| "grad_norm": 0.007915050722658634, | |
| "learning_rate": 5.344827586206896e-05, | |
| "loss": 0.0016, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.9655172413793105, | |
| "grad_norm": 0.008189872838556767, | |
| "learning_rate": 5.172413793103449e-05, | |
| "loss": 0.0016, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.007728048134595156, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0016, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.0344827586206895, | |
| "grad_norm": 0.007373048458248377, | |
| "learning_rate": 4.827586206896552e-05, | |
| "loss": 0.0016, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.0689655172413794, | |
| "grad_norm": 0.00832042470574379, | |
| "learning_rate": 4.655172413793104e-05, | |
| "loss": 0.0016, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 3.103448275862069, | |
| "grad_norm": 0.007547506596893072, | |
| "learning_rate": 4.482758620689655e-05, | |
| "loss": 0.0016, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.103448275862069, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0015662283403798938, | |
| "eval_runtime": 9.1484, | |
| "eval_samples_per_second": 56.513, | |
| "eval_steps_per_second": 7.105, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.1379310344827585, | |
| "grad_norm": 0.007507437374442816, | |
| "learning_rate": 4.3103448275862066e-05, | |
| "loss": 0.0016, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.1724137931034484, | |
| "grad_norm": 0.008085578680038452, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "loss": 0.0015, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.206896551724138, | |
| "grad_norm": 0.007526910398155451, | |
| "learning_rate": 3.965517241379311e-05, | |
| "loss": 0.0015, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.2413793103448274, | |
| "grad_norm": 0.007942083291709423, | |
| "learning_rate": 3.793103448275862e-05, | |
| "loss": 0.0015, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.2758620689655173, | |
| "grad_norm": 0.007609126623719931, | |
| "learning_rate": 3.620689655172414e-05, | |
| "loss": 0.0015, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.310344827586207, | |
| "grad_norm": 0.007188369985669851, | |
| "learning_rate": 3.4482758620689657e-05, | |
| "loss": 0.0015, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.344827586206897, | |
| "grad_norm": 0.0077166566625237465, | |
| "learning_rate": 3.275862068965517e-05, | |
| "loss": 0.0015, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.3793103448275863, | |
| "grad_norm": 0.007557329256087542, | |
| "learning_rate": 3.103448275862069e-05, | |
| "loss": 0.0015, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.413793103448276, | |
| "grad_norm": 0.006876759696751833, | |
| "learning_rate": 2.9310344827586206e-05, | |
| "loss": 0.0015, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.4482758620689653, | |
| "grad_norm": 0.007274131290614605, | |
| "learning_rate": 2.7586206896551727e-05, | |
| "loss": 0.0015, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.4482758620689653, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.001456769765354693, | |
| "eval_runtime": 9.1926, | |
| "eval_samples_per_second": 56.241, | |
| "eval_steps_per_second": 7.071, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.4827586206896552, | |
| "grad_norm": 0.007954578846693039, | |
| "learning_rate": 2.5862068965517244e-05, | |
| "loss": 0.0015, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.5172413793103448, | |
| "grad_norm": 0.007275938987731934, | |
| "learning_rate": 2.413793103448276e-05, | |
| "loss": 0.0015, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.5517241379310347, | |
| "grad_norm": 0.006545715499669313, | |
| "learning_rate": 2.2413793103448276e-05, | |
| "loss": 0.0014, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.586206896551724, | |
| "grad_norm": 0.006517117377370596, | |
| "learning_rate": 2.0689655172413793e-05, | |
| "loss": 0.0014, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.6206896551724137, | |
| "grad_norm": 0.006700526457279921, | |
| "learning_rate": 1.896551724137931e-05, | |
| "loss": 0.0014, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.655172413793103, | |
| "grad_norm": 0.006489131134003401, | |
| "learning_rate": 1.7241379310344828e-05, | |
| "loss": 0.0014, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.689655172413793, | |
| "grad_norm": 0.006455055437982082, | |
| "learning_rate": 1.5517241379310346e-05, | |
| "loss": 0.0014, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.7241379310344827, | |
| "grad_norm": 0.006427043117582798, | |
| "learning_rate": 1.3793103448275863e-05, | |
| "loss": 0.0014, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.7586206896551726, | |
| "grad_norm": 0.006561470218002796, | |
| "learning_rate": 1.206896551724138e-05, | |
| "loss": 0.0014, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.793103448275862, | |
| "grad_norm": 0.007191502954810858, | |
| "learning_rate": 1.0344827586206897e-05, | |
| "loss": 0.0014, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.793103448275862, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0014008664293214679, | |
| "eval_runtime": 9.2274, | |
| "eval_samples_per_second": 56.029, | |
| "eval_steps_per_second": 7.044, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.8275862068965516, | |
| "grad_norm": 0.006780336145311594, | |
| "learning_rate": 8.620689655172414e-06, | |
| "loss": 0.0014, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.862068965517241, | |
| "grad_norm": 0.006570115685462952, | |
| "learning_rate": 6.896551724137932e-06, | |
| "loss": 0.0014, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.896551724137931, | |
| "grad_norm": 0.00782351940870285, | |
| "learning_rate": 5.172413793103448e-06, | |
| "loss": 0.0014, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.9310344827586206, | |
| "grad_norm": 0.006399817299097776, | |
| "learning_rate": 3.448275862068966e-06, | |
| "loss": 0.0014, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.9655172413793105, | |
| "grad_norm": 0.00720432261005044, | |
| "learning_rate": 1.724137931034483e-06, | |
| "loss": 0.0014, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.006754900328814983, | |
| "learning_rate": 0.0, | |
| "loss": 0.0014, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1160, | |
| "total_flos": 1.436727240856535e+18, | |
| "train_loss": 0.027367618223170524, | |
| "train_runtime": 673.711, | |
| "train_samples_per_second": 27.519, | |
| "train_steps_per_second": 1.722 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.436727240856535e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |