| { | |
| "best_metric": 0.9844, | |
| "best_model_checkpoint": "vit-small-patch16-224-finetuned-cifar10/checkpoint-1170", | |
| "epoch": 2.9942418426103647, | |
| "eval_steps": 500, | |
| "global_step": 1170, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 8.970051765441895, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 2.432, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 7.947530746459961, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 2.2376, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 7.60875129699707, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 1.9698, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 6.434678554534912, | |
| "learning_rate": 1.7094017094017095e-05, | |
| "loss": 1.5489, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 6.238937854766846, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 1.1375, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 5.423764228820801, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 0.7332, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 7.145791530609131, | |
| "learning_rate": 2.9914529914529915e-05, | |
| "loss": 0.5841, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 6.430414199829102, | |
| "learning_rate": 3.418803418803419e-05, | |
| "loss": 0.4957, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 6.975243091583252, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.4523, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 6.09722900390625, | |
| "learning_rate": 4.2735042735042735e-05, | |
| "loss": 0.4462, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 5.052177429199219, | |
| "learning_rate": 4.700854700854701e-05, | |
| "loss": 0.3624, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 5.480886459350586, | |
| "learning_rate": 4.985754985754986e-05, | |
| "loss": 0.3895, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 4.691812992095947, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 0.3446, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 6.404294013977051, | |
| "learning_rate": 4.890788224121557e-05, | |
| "loss": 0.4006, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 5.33477258682251, | |
| "learning_rate": 4.8433048433048433e-05, | |
| "loss": 0.3532, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 5.476822853088379, | |
| "learning_rate": 4.7958214624881294e-05, | |
| "loss": 0.3421, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 6.005227565765381, | |
| "learning_rate": 4.7483380816714154e-05, | |
| "loss": 0.3692, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 4.84128999710083, | |
| "learning_rate": 4.700854700854701e-05, | |
| "loss": 0.3457, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 4.650127410888672, | |
| "learning_rate": 4.653371320037987e-05, | |
| "loss": 0.3685, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 5.779717445373535, | |
| "learning_rate": 4.605887939221273e-05, | |
| "loss": 0.3451, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 6.38178825378418, | |
| "learning_rate": 4.558404558404559e-05, | |
| "loss": 0.3394, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 5.148958206176758, | |
| "learning_rate": 4.510921177587845e-05, | |
| "loss": 0.3641, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 3.9889109134674072, | |
| "learning_rate": 4.463437796771131e-05, | |
| "loss": 0.3108, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 6.51384162902832, | |
| "learning_rate": 4.415954415954416e-05, | |
| "loss": 0.2949, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 4.950154781341553, | |
| "learning_rate": 4.368471035137702e-05, | |
| "loss": 0.3352, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 6.234986782073975, | |
| "learning_rate": 4.3209876543209875e-05, | |
| "loss": 0.309, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 5.57274055480957, | |
| "learning_rate": 4.2735042735042735e-05, | |
| "loss": 0.3161, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 5.511316776275635, | |
| "learning_rate": 4.2260208926875595e-05, | |
| "loss": 0.3018, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 3.934835910797119, | |
| "learning_rate": 4.1785375118708455e-05, | |
| "loss": 0.3116, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 5.323722839355469, | |
| "learning_rate": 4.131054131054131e-05, | |
| "loss": 0.3105, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 4.1425909996032715, | |
| "learning_rate": 4.083570750237417e-05, | |
| "loss": 0.296, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 5.026642799377441, | |
| "learning_rate": 4.036087369420703e-05, | |
| "loss": 0.3594, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 3.949415922164917, | |
| "learning_rate": 3.988603988603989e-05, | |
| "loss": 0.2849, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 4.380434989929199, | |
| "learning_rate": 3.941120607787275e-05, | |
| "loss": 0.2935, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 4.884699821472168, | |
| "learning_rate": 3.893637226970561e-05, | |
| "loss": 0.3128, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 6.950473308563232, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.3073, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 5.410139560699463, | |
| "learning_rate": 3.798670465337132e-05, | |
| "loss": 0.2998, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 4.638174057006836, | |
| "learning_rate": 3.7511870845204176e-05, | |
| "loss": 0.3124, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.711712837219238, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2682, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9713, | |
| "eval_loss": 0.0821285992860794, | |
| "eval_runtime": 32.3489, | |
| "eval_samples_per_second": 309.13, | |
| "eval_steps_per_second": 9.676, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 4.979706764221191, | |
| "learning_rate": 3.65622032288699e-05, | |
| "loss": 0.2836, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 3.807408332824707, | |
| "learning_rate": 3.608736942070276e-05, | |
| "loss": 0.2521, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 4.402036190032959, | |
| "learning_rate": 3.561253561253561e-05, | |
| "loss": 0.2469, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.0667805671691895, | |
| "learning_rate": 3.513770180436847e-05, | |
| "loss": 0.2383, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 6.403406620025635, | |
| "learning_rate": 3.466286799620133e-05, | |
| "loss": 0.277, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 5.481479167938232, | |
| "learning_rate": 3.418803418803419e-05, | |
| "loss": 0.2533, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 5.324038028717041, | |
| "learning_rate": 3.371320037986705e-05, | |
| "loss": 0.2813, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.4851675033569336, | |
| "learning_rate": 3.323836657169991e-05, | |
| "loss": 0.3055, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 4.365708827972412, | |
| "learning_rate": 3.2763532763532764e-05, | |
| "loss": 0.2797, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 3.6004509925842285, | |
| "learning_rate": 3.2288698955365625e-05, | |
| "loss": 0.2776, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 3.7064690589904785, | |
| "learning_rate": 3.181386514719848e-05, | |
| "loss": 0.2605, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 4.883594036102295, | |
| "learning_rate": 3.133903133903134e-05, | |
| "loss": 0.2403, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 5.014550685882568, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 0.2436, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 4.304734230041504, | |
| "learning_rate": 3.0389363722697055e-05, | |
| "loss": 0.2639, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 3.2759273052215576, | |
| "learning_rate": 2.9914529914529915e-05, | |
| "loss": 0.2651, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 3.556528091430664, | |
| "learning_rate": 2.9439696106362775e-05, | |
| "loss": 0.2359, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 4.672804355621338, | |
| "learning_rate": 2.8964862298195632e-05, | |
| "loss": 0.2376, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 4.527768611907959, | |
| "learning_rate": 2.8490028490028492e-05, | |
| "loss": 0.2504, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 5.102674961090088, | |
| "learning_rate": 2.8015194681861352e-05, | |
| "loss": 0.2604, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 4.18637228012085, | |
| "learning_rate": 2.754036087369421e-05, | |
| "loss": 0.2746, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 3.3811545372009277, | |
| "learning_rate": 2.706552706552707e-05, | |
| "loss": 0.2548, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 3.1672608852386475, | |
| "learning_rate": 2.6590693257359926e-05, | |
| "loss": 0.2423, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 3.609534740447998, | |
| "learning_rate": 2.611585944919278e-05, | |
| "loss": 0.1994, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 5.083642482757568, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 0.292, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 4.716630935668945, | |
| "learning_rate": 2.51661918328585e-05, | |
| "loss": 0.2333, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 4.915971755981445, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.2364, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 4.216696739196777, | |
| "learning_rate": 2.4216524216524217e-05, | |
| "loss": 0.2503, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 4.966453552246094, | |
| "learning_rate": 2.3741690408357077e-05, | |
| "loss": 0.224, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 4.153652191162109, | |
| "learning_rate": 2.3266856600189934e-05, | |
| "loss": 0.2109, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 2.9919214248657227, | |
| "learning_rate": 2.2792022792022794e-05, | |
| "loss": 0.2371, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 5.105522155761719, | |
| "learning_rate": 2.2317188983855654e-05, | |
| "loss": 0.2187, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 3.9936702251434326, | |
| "learning_rate": 2.184235517568851e-05, | |
| "loss": 0.2403, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 5.220417022705078, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 0.2383, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 3.134110450744629, | |
| "learning_rate": 2.0892687559354228e-05, | |
| "loss": 0.2197, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 5.406284809112549, | |
| "learning_rate": 2.0417853751187084e-05, | |
| "loss": 0.224, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 5.273613452911377, | |
| "learning_rate": 1.9943019943019945e-05, | |
| "loss": 0.1951, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 4.695704936981201, | |
| "learning_rate": 1.9468186134852805e-05, | |
| "loss": 0.2324, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "grad_norm": 5.734136581420898, | |
| "learning_rate": 1.899335232668566e-05, | |
| "loss": 0.248, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.757730960845947, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.252, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9842, | |
| "eval_loss": 0.050203289836645126, | |
| "eval_runtime": 31.7681, | |
| "eval_samples_per_second": 314.781, | |
| "eval_steps_per_second": 9.853, | |
| "step": 781 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 4.400415897369385, | |
| "learning_rate": 1.804368471035138e-05, | |
| "loss": 0.2364, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 4.02318000793457, | |
| "learning_rate": 1.7568850902184235e-05, | |
| "loss": 0.2164, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 5.289691925048828, | |
| "learning_rate": 1.7094017094017095e-05, | |
| "loss": 0.2472, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 3.842559337615967, | |
| "learning_rate": 1.6619183285849956e-05, | |
| "loss": 0.2209, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 5.64124870300293, | |
| "learning_rate": 1.6144349477682812e-05, | |
| "loss": 0.2473, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 4.473005771636963, | |
| "learning_rate": 1.566951566951567e-05, | |
| "loss": 0.2287, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 4.476832389831543, | |
| "learning_rate": 1.5194681861348528e-05, | |
| "loss": 0.2208, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 4.89029598236084, | |
| "learning_rate": 1.4719848053181388e-05, | |
| "loss": 0.2332, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 5.0262627601623535, | |
| "learning_rate": 1.4245014245014246e-05, | |
| "loss": 0.2001, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 4.80778169631958, | |
| "learning_rate": 1.3770180436847105e-05, | |
| "loss": 0.2018, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 3.203700304031372, | |
| "learning_rate": 1.3295346628679963e-05, | |
| "loss": 0.2018, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 4.840007305145264, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 0.2754, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 4.501500129699707, | |
| "learning_rate": 1.2345679012345678e-05, | |
| "loss": 0.1986, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 3.4367220401763916, | |
| "learning_rate": 1.1870845204178538e-05, | |
| "loss": 0.2145, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 5.076148509979248, | |
| "learning_rate": 1.1396011396011397e-05, | |
| "loss": 0.2001, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 2.299694299697876, | |
| "learning_rate": 1.0921177587844255e-05, | |
| "loss": 0.1935, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 3.9149887561798096, | |
| "learning_rate": 1.0446343779677114e-05, | |
| "loss": 0.2158, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 5.087822437286377, | |
| "learning_rate": 9.971509971509972e-06, | |
| "loss": 0.2052, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 3.8664252758026123, | |
| "learning_rate": 9.49667616334283e-06, | |
| "loss": 0.2056, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 4.961308479309082, | |
| "learning_rate": 9.02184235517569e-06, | |
| "loss": 0.2157, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 4.553085803985596, | |
| "learning_rate": 8.547008547008548e-06, | |
| "loss": 0.2079, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 3.396073818206787, | |
| "learning_rate": 8.072174738841406e-06, | |
| "loss": 0.1919, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 5.101813316345215, | |
| "learning_rate": 7.597340930674264e-06, | |
| "loss": 0.1812, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 5.189543724060059, | |
| "learning_rate": 7.122507122507123e-06, | |
| "loss": 0.2123, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 4.240951061248779, | |
| "learning_rate": 6.6476733143399815e-06, | |
| "loss": 0.218, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 3.139678716659546, | |
| "learning_rate": 6.172839506172839e-06, | |
| "loss": 0.1824, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 2.84495210647583, | |
| "learning_rate": 5.6980056980056985e-06, | |
| "loss": 0.1916, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 3.4231679439544678, | |
| "learning_rate": 5.223171889838557e-06, | |
| "loss": 0.1953, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 4.120250225067139, | |
| "learning_rate": 4.748338081671415e-06, | |
| "loss": 0.2046, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 5.0515055656433105, | |
| "learning_rate": 4.273504273504274e-06, | |
| "loss": 0.1941, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 4.197494029998779, | |
| "learning_rate": 3.798670465337132e-06, | |
| "loss": 0.1908, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 5.316411972045898, | |
| "learning_rate": 3.3238366571699908e-06, | |
| "loss": 0.1954, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 2.8527472019195557, | |
| "learning_rate": 2.8490028490028492e-06, | |
| "loss": 0.1466, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 4.074756622314453, | |
| "learning_rate": 2.3741690408357077e-06, | |
| "loss": 0.2092, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 4.336794853210449, | |
| "learning_rate": 1.899335232668566e-06, | |
| "loss": 0.1742, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 3.8528594970703125, | |
| "learning_rate": 1.4245014245014246e-06, | |
| "loss": 0.2002, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 4.159782409667969, | |
| "learning_rate": 9.49667616334283e-07, | |
| "loss": 0.1847, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 5.317773342132568, | |
| "learning_rate": 4.748338081671415e-07, | |
| "loss": 0.2293, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 4.241806983947754, | |
| "learning_rate": 0.0, | |
| "loss": 0.173, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.9844, | |
| "eval_loss": 0.04590694606304169, | |
| "eval_runtime": 32.5039, | |
| "eval_samples_per_second": 307.655, | |
| "eval_steps_per_second": 9.63, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "step": 1170, | |
| "total_flos": 2.930358373492064e+18, | |
| "train_loss": 0.33610059461023056, | |
| "train_runtime": 935.657, | |
| "train_samples_per_second": 160.315, | |
| "train_steps_per_second": 1.25 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1170, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.930358373492064e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |