| { | |
| "best_metric": 0.5187094211578369, | |
| "best_model_checkpoint": "./vit-lr-0.0001/checkpoint-800", | |
| "epoch": 3.426791277258567, | |
| "eval_steps": 100, | |
| "global_step": 1100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 3.3859810829162598, | |
| "learning_rate": 9.997196261682243e-05, | |
| "loss": 1.1978, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 4.35668420791626, | |
| "learning_rate": 9.994080996884736e-05, | |
| "loss": 0.717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 5.1708455085754395, | |
| "learning_rate": 9.990965732087227e-05, | |
| "loss": 0.8267, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 9.000005722045898, | |
| "learning_rate": 9.98785046728972e-05, | |
| "loss": 1.0157, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 4.164905071258545, | |
| "learning_rate": 9.984735202492212e-05, | |
| "loss": 0.7406, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 3.7239935398101807, | |
| "learning_rate": 9.981619937694705e-05, | |
| "loss": 0.7899, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 3.9820024967193604, | |
| "learning_rate": 9.978504672897196e-05, | |
| "loss": 0.816, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 3.7813899517059326, | |
| "learning_rate": 9.975389408099689e-05, | |
| "loss": 0.6441, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 6.419355869293213, | |
| "learning_rate": 9.972274143302182e-05, | |
| "loss": 0.6568, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 5.270507335662842, | |
| "learning_rate": 9.969158878504672e-05, | |
| "loss": 0.6008, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.7836338418862691, | |
| "eval_loss": 0.5998427271842957, | |
| "eval_runtime": 39.7401, | |
| "eval_samples_per_second": 72.571, | |
| "eval_steps_per_second": 9.084, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 4.352388858795166, | |
| "learning_rate": 9.966043613707165e-05, | |
| "loss": 0.6567, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 6.960423946380615, | |
| "learning_rate": 9.962928348909658e-05, | |
| "loss": 0.643, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 8.54637336730957, | |
| "learning_rate": 9.95981308411215e-05, | |
| "loss": 0.6606, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.6916728019714355, | |
| "learning_rate": 9.956697819314643e-05, | |
| "loss": 0.5345, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 5.490638732910156, | |
| "learning_rate": 9.953582554517134e-05, | |
| "loss": 0.6008, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 6.159054279327393, | |
| "learning_rate": 9.950467289719627e-05, | |
| "loss": 0.5466, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 2.6157217025756836, | |
| "learning_rate": 9.947352024922119e-05, | |
| "loss": 0.5968, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 7.270638465881348, | |
| "learning_rate": 9.944236760124612e-05, | |
| "loss": 0.6522, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 6.296653747558594, | |
| "learning_rate": 9.941121495327103e-05, | |
| "loss": 0.5591, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 6.226278781890869, | |
| "learning_rate": 9.938006230529595e-05, | |
| "loss": 0.5779, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_accuracy": 0.7832871012482663, | |
| "eval_loss": 0.611842930316925, | |
| "eval_runtime": 39.418, | |
| "eval_samples_per_second": 73.165, | |
| "eval_steps_per_second": 9.158, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 4.321172714233398, | |
| "learning_rate": 9.934890965732088e-05, | |
| "loss": 0.606, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 6.226727485656738, | |
| "learning_rate": 9.93177570093458e-05, | |
| "loss": 0.7798, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 3.879312515258789, | |
| "learning_rate": 9.928660436137072e-05, | |
| "loss": 0.4047, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.546088457107544, | |
| "learning_rate": 9.925545171339564e-05, | |
| "loss": 0.559, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 6.213643550872803, | |
| "learning_rate": 9.922429906542056e-05, | |
| "loss": 0.7639, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 3.789020299911499, | |
| "learning_rate": 9.91931464174455e-05, | |
| "loss": 0.5372, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 5.764331340789795, | |
| "learning_rate": 9.916199376947041e-05, | |
| "loss": 0.5031, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 2.2293648719787598, | |
| "learning_rate": 9.913084112149534e-05, | |
| "loss": 0.4379, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 6.797378063201904, | |
| "learning_rate": 9.909968847352025e-05, | |
| "loss": 0.6147, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 6.964852809906006, | |
| "learning_rate": 9.906853582554517e-05, | |
| "loss": 0.6891, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_accuracy": 0.7777392510402219, | |
| "eval_loss": 0.5814183354377747, | |
| "eval_runtime": 39.6347, | |
| "eval_samples_per_second": 72.765, | |
| "eval_steps_per_second": 9.108, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 3.367398262023926, | |
| "learning_rate": 9.90373831775701e-05, | |
| "loss": 0.5153, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.517517566680908, | |
| "learning_rate": 9.900623052959503e-05, | |
| "loss": 0.5022, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 6.2498393058776855, | |
| "learning_rate": 9.897507788161994e-05, | |
| "loss": 0.4581, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 5.515881538391113, | |
| "learning_rate": 9.894392523364486e-05, | |
| "loss": 0.3453, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 3.255171060562134, | |
| "learning_rate": 9.891277258566979e-05, | |
| "loss": 0.4332, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 2.262460231781006, | |
| "learning_rate": 9.888161993769472e-05, | |
| "loss": 0.3463, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 5.0724711418151855, | |
| "learning_rate": 9.885046728971963e-05, | |
| "loss": 0.5136, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 3.8894565105438232, | |
| "learning_rate": 9.881931464174455e-05, | |
| "loss": 0.3792, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 4.299795627593994, | |
| "learning_rate": 9.878816199376948e-05, | |
| "loss": 0.4804, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.381559371948242, | |
| "learning_rate": 9.875700934579439e-05, | |
| "loss": 0.3535, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.7933425797503467, | |
| "eval_loss": 0.5594320297241211, | |
| "eval_runtime": 39.1416, | |
| "eval_samples_per_second": 73.681, | |
| "eval_steps_per_second": 9.223, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 4.152671813964844, | |
| "learning_rate": 9.872585669781932e-05, | |
| "loss": 0.4629, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 3.4808621406555176, | |
| "learning_rate": 9.869470404984425e-05, | |
| "loss": 0.3583, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 1.483371615409851, | |
| "learning_rate": 9.866355140186917e-05, | |
| "loss": 0.5018, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 4.952483654022217, | |
| "learning_rate": 9.863239875389408e-05, | |
| "loss": 0.4338, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 4.527029037475586, | |
| "learning_rate": 9.860124610591901e-05, | |
| "loss": 0.437, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 7.609699726104736, | |
| "learning_rate": 9.857009345794394e-05, | |
| "loss": 0.3796, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 3.572444200515747, | |
| "learning_rate": 9.853894080996885e-05, | |
| "loss": 0.3551, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 8.147821426391602, | |
| "learning_rate": 9.850778816199377e-05, | |
| "loss": 0.4182, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 6.7267279624938965, | |
| "learning_rate": 9.84766355140187e-05, | |
| "loss": 0.3584, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 3.457125425338745, | |
| "learning_rate": 9.844548286604361e-05, | |
| "loss": 0.4055, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.8134535367545076, | |
| "eval_loss": 0.571877658367157, | |
| "eval_runtime": 39.3357, | |
| "eval_samples_per_second": 73.318, | |
| "eval_steps_per_second": 9.177, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 4.186713695526123, | |
| "learning_rate": 9.841433021806854e-05, | |
| "loss": 0.4802, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 4.469476699829102, | |
| "learning_rate": 9.838317757009346e-05, | |
| "loss": 0.4118, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 7.559543609619141, | |
| "learning_rate": 9.835202492211837e-05, | |
| "loss": 0.4605, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 2.5254476070404053, | |
| "learning_rate": 9.83208722741433e-05, | |
| "loss": 0.2878, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 4.755306720733643, | |
| "learning_rate": 9.828971962616823e-05, | |
| "loss": 0.3881, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 3.8046302795410156, | |
| "learning_rate": 9.825856697819316e-05, | |
| "loss": 0.3511, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 2.603055238723755, | |
| "learning_rate": 9.822741433021808e-05, | |
| "loss": 0.3354, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 3.5978047847747803, | |
| "learning_rate": 9.819626168224299e-05, | |
| "loss": 0.2885, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 5.625058174133301, | |
| "learning_rate": 9.816510903426792e-05, | |
| "loss": 0.392, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 4.636626720428467, | |
| "learning_rate": 9.813395638629284e-05, | |
| "loss": 0.2471, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.794382801664355, | |
| "eval_loss": 0.5536235570907593, | |
| "eval_runtime": 39.2047, | |
| "eval_samples_per_second": 73.563, | |
| "eval_steps_per_second": 9.208, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 6.1720123291015625, | |
| "learning_rate": 9.810280373831777e-05, | |
| "loss": 0.5169, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 3.5033581256866455, | |
| "learning_rate": 9.807165109034268e-05, | |
| "loss": 0.4374, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 4.63729190826416, | |
| "learning_rate": 9.80404984423676e-05, | |
| "loss": 0.2866, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 3.4656624794006348, | |
| "learning_rate": 9.800934579439253e-05, | |
| "loss": 0.3323, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 3.2712185382843018, | |
| "learning_rate": 9.797819314641746e-05, | |
| "loss": 0.2275, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 3.9931576251983643, | |
| "learning_rate": 9.794704049844237e-05, | |
| "loss": 0.21, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 1.7855778932571411, | |
| "learning_rate": 9.791588785046729e-05, | |
| "loss": 0.2996, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 5.909547328948975, | |
| "learning_rate": 9.788473520249222e-05, | |
| "loss": 0.2081, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 1.9149627685546875, | |
| "learning_rate": 9.785358255451714e-05, | |
| "loss": 0.216, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 4.954345226287842, | |
| "learning_rate": 9.782242990654206e-05, | |
| "loss": 0.171, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_accuracy": 0.8391123439667129, | |
| "eval_loss": 0.5375762581825256, | |
| "eval_runtime": 39.0099, | |
| "eval_samples_per_second": 73.93, | |
| "eval_steps_per_second": 9.254, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 1.2202866077423096, | |
| "learning_rate": 9.779127725856699e-05, | |
| "loss": 0.1488, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 6.487290382385254, | |
| "learning_rate": 9.77601246105919e-05, | |
| "loss": 0.3093, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 6.3305230140686035, | |
| "learning_rate": 9.772897196261682e-05, | |
| "loss": 0.3298, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 1.8739694356918335, | |
| "learning_rate": 9.769781931464175e-05, | |
| "loss": 0.2836, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 6.215338230133057, | |
| "learning_rate": 9.766666666666668e-05, | |
| "loss": 0.213, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 0.6346792578697205, | |
| "learning_rate": 9.76355140186916e-05, | |
| "loss": 0.1537, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 6.488032341003418, | |
| "learning_rate": 9.760436137071651e-05, | |
| "loss": 0.2713, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 3.9325110912323, | |
| "learning_rate": 9.757320872274144e-05, | |
| "loss": 0.2861, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 4.505757808685303, | |
| "learning_rate": 9.754205607476637e-05, | |
| "loss": 0.2625, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 1.1799720525741577, | |
| "learning_rate": 9.751090342679128e-05, | |
| "loss": 0.2159, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_accuracy": 0.8359916782246879, | |
| "eval_loss": 0.5187094211578369, | |
| "eval_runtime": 38.9237, | |
| "eval_samples_per_second": 74.094, | |
| "eval_steps_per_second": 9.275, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 0.662136971950531, | |
| "learning_rate": 9.74797507788162e-05, | |
| "loss": 0.2325, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 9.680724143981934, | |
| "learning_rate": 9.744859813084113e-05, | |
| "loss": 0.2579, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 0.5741639137268066, | |
| "learning_rate": 9.741744548286604e-05, | |
| "loss": 0.1271, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 9.787598609924316, | |
| "learning_rate": 9.738629283489097e-05, | |
| "loss": 0.2705, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 3.5786948204040527, | |
| "learning_rate": 9.73551401869159e-05, | |
| "loss": 0.1892, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 3.534965753555298, | |
| "learning_rate": 9.732398753894082e-05, | |
| "loss": 0.1629, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 3.6333532333374023, | |
| "learning_rate": 9.729283489096573e-05, | |
| "loss": 0.3307, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 7.089163780212402, | |
| "learning_rate": 9.726168224299066e-05, | |
| "loss": 0.3159, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "grad_norm": 0.845410168170929, | |
| "learning_rate": 9.723052959501559e-05, | |
| "loss": 0.1572, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 6.727290630340576, | |
| "learning_rate": 9.71993769470405e-05, | |
| "loss": 0.1659, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_accuracy": 0.8318307905686546, | |
| "eval_loss": 0.5848782062530518, | |
| "eval_runtime": 38.908, | |
| "eval_samples_per_second": 74.124, | |
| "eval_steps_per_second": 9.278, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 4.609592914581299, | |
| "learning_rate": 9.716822429906542e-05, | |
| "loss": 0.3052, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 4.7995285987854, | |
| "learning_rate": 9.713707165109035e-05, | |
| "loss": 0.3944, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 4.78571891784668, | |
| "learning_rate": 9.710591900311527e-05, | |
| "loss": 0.3118, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "grad_norm": 7.084194183349609, | |
| "learning_rate": 9.70747663551402e-05, | |
| "loss": 0.2874, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 2.6003012657165527, | |
| "learning_rate": 9.704361370716511e-05, | |
| "loss": 0.2024, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 3.7201614379882812, | |
| "learning_rate": 9.701246105919004e-05, | |
| "loss": 0.224, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 3.0651230812072754, | |
| "learning_rate": 9.698130841121495e-05, | |
| "loss": 0.1076, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 4.620060920715332, | |
| "learning_rate": 9.695015576323988e-05, | |
| "loss": 0.1312, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 0.5105383992195129, | |
| "learning_rate": 9.691900311526481e-05, | |
| "loss": 0.1479, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 1.0817281007766724, | |
| "learning_rate": 9.688785046728971e-05, | |
| "loss": 0.0554, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.8543689320388349, | |
| "eval_loss": 0.5457024574279785, | |
| "eval_runtime": 39.7583, | |
| "eval_samples_per_second": 72.538, | |
| "eval_steps_per_second": 9.08, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 1.351655125617981, | |
| "learning_rate": 9.685669781931464e-05, | |
| "loss": 0.0568, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "grad_norm": 3.9608170986175537, | |
| "learning_rate": 9.682554517133957e-05, | |
| "loss": 0.1162, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "grad_norm": 6.9148783683776855, | |
| "learning_rate": 9.679439252336449e-05, | |
| "loss": 0.0587, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "grad_norm": 1.0946722030639648, | |
| "learning_rate": 9.676323987538942e-05, | |
| "loss": 0.0572, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 9.527204513549805, | |
| "learning_rate": 9.673208722741433e-05, | |
| "loss": 0.1847, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 7.0584716796875, | |
| "learning_rate": 9.670093457943926e-05, | |
| "loss": 0.1296, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 7.145915985107422, | |
| "learning_rate": 9.666978193146418e-05, | |
| "loss": 0.1188, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 7.105827808380127, | |
| "learning_rate": 9.66386292834891e-05, | |
| "loss": 0.099, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 8.103167533874512, | |
| "learning_rate": 9.660747663551402e-05, | |
| "loss": 0.0874, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "grad_norm": 1.4485965967178345, | |
| "learning_rate": 9.657632398753894e-05, | |
| "loss": 0.1125, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "eval_accuracy": 0.8318307905686546, | |
| "eval_loss": 0.7323819398880005, | |
| "eval_runtime": 39.6414, | |
| "eval_samples_per_second": 72.752, | |
| "eval_steps_per_second": 9.107, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "step": 1100, | |
| "total_flos": 1.362060245306622e+18, | |
| "train_loss": 0.38896639000285754, | |
| "train_runtime": 855.3091, | |
| "train_samples_per_second": 599.549, | |
| "train_steps_per_second": 37.53 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 32100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 100, | |
| "total_flos": 1.362060245306622e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |