| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 9822, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.9898187741804116e-05, | |
| "loss": 0.5828, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.9796375483608227e-05, | |
| "loss": 0.4999, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.9694563225412342e-05, | |
| "loss": 0.4401, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.9592750967216453e-05, | |
| "loss": 0.4643, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.9490938709020567e-05, | |
| "loss": 0.4309, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.938912645082468e-05, | |
| "loss": 0.4221, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.9287314192628793e-05, | |
| "loss": 0.4139, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.9185501934432908e-05, | |
| "loss": 0.4245, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9083689676237022e-05, | |
| "loss": 0.4345, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.8981877418041133e-05, | |
| "loss": 0.4041, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.8488010250777961, | |
| "eval_loss": 0.36110368371009827, | |
| "eval_runtime": 10.7308, | |
| "eval_samples_per_second": 509.095, | |
| "eval_steps_per_second": 31.871, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.8880065159845248e-05, | |
| "loss": 0.4295, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.8778252901649362e-05, | |
| "loss": 0.388, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.8676440643453473e-05, | |
| "loss": 0.3871, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.8574628385257588e-05, | |
| "loss": 0.4007, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.84728161270617e-05, | |
| "loss": 0.3881, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.8371003868865813e-05, | |
| "loss": 0.3601, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.8269191610669924e-05, | |
| "loss": 0.4022, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.816737935247404e-05, | |
| "loss": 0.3937, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.8065567094278153e-05, | |
| "loss": 0.3675, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.7963754836082265e-05, | |
| "loss": 0.3784, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.8603331502837269, | |
| "eval_loss": 0.32318630814552307, | |
| "eval_runtime": 10.3027, | |
| "eval_samples_per_second": 530.251, | |
| "eval_steps_per_second": 33.195, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.786194257788638e-05, | |
| "loss": 0.3774, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.776013031969049e-05, | |
| "loss": 0.4018, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1.7658318061494605e-05, | |
| "loss": 0.3858, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.755650580329872e-05, | |
| "loss": 0.3978, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.7454693545102834e-05, | |
| "loss": 0.36, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.7352881286906945e-05, | |
| "loss": 0.3697, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.725106902871106e-05, | |
| "loss": 0.3562, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.7149256770515174e-05, | |
| "loss": 0.3963, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.7047444512319285e-05, | |
| "loss": 0.3659, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.69456322541234e-05, | |
| "loss": 0.364, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_accuracy": 0.8641771920190372, | |
| "eval_loss": 0.3128460645675659, | |
| "eval_runtime": 10.3115, | |
| "eval_samples_per_second": 529.798, | |
| "eval_steps_per_second": 33.167, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.684381999592751e-05, | |
| "loss": 0.3548, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.6742007737731625e-05, | |
| "loss": 0.3479, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.6640195479535736e-05, | |
| "loss": 0.3439, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.653838322133985e-05, | |
| "loss": 0.3501, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1.6436570963143962e-05, | |
| "loss": 0.3573, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.6334758704948076e-05, | |
| "loss": 0.3639, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.623294644675219e-05, | |
| "loss": 0.3581, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.6131134188556302e-05, | |
| "loss": 0.3354, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.6029321930360416e-05, | |
| "loss": 0.3309, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.592750967216453e-05, | |
| "loss": 0.364, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.8702178290316676, | |
| "eval_loss": 0.3019617199897766, | |
| "eval_runtime": 10.3378, | |
| "eval_samples_per_second": 528.449, | |
| "eval_steps_per_second": 33.082, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.5825697413968642e-05, | |
| "loss": 0.3361, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.5723885155772757e-05, | |
| "loss": 0.3388, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.562207289757687e-05, | |
| "loss": 0.3222, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.5520260639380982e-05, | |
| "loss": 0.3205, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.5418448381185097e-05, | |
| "loss": 0.3414, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.531663612298921e-05, | |
| "loss": 0.3542, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.5214823864793322e-05, | |
| "loss": 0.3404, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.5113011606597437e-05, | |
| "loss": 0.3518, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.5011199348401548e-05, | |
| "loss": 0.3503, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.4909387090205662e-05, | |
| "loss": 0.3236, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.876807614863628, | |
| "eval_loss": 0.2959628701210022, | |
| "eval_runtime": 10.7839, | |
| "eval_samples_per_second": 506.587, | |
| "eval_steps_per_second": 31.714, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.4807574832009775e-05, | |
| "loss": 0.352, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.4705762573813888e-05, | |
| "loss": 0.337, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.4603950315618001e-05, | |
| "loss": 0.325, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.4502138057422115e-05, | |
| "loss": 0.3505, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.440032579922623e-05, | |
| "loss": 0.306, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.4298513541030341e-05, | |
| "loss": 0.3491, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.4196701282834456e-05, | |
| "loss": 0.3223, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.4094889024638567e-05, | |
| "loss": 0.3366, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.3993076766442681e-05, | |
| "loss": 0.3215, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.3891264508246794e-05, | |
| "loss": 0.3475, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.8815669046311551, | |
| "eval_loss": 0.28946802020072937, | |
| "eval_runtime": 10.2811, | |
| "eval_samples_per_second": 531.361, | |
| "eval_steps_per_second": 33.265, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.3789452250050907e-05, | |
| "loss": 0.3258, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.368763999185502e-05, | |
| "loss": 0.344, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.3585827733659134e-05, | |
| "loss": 0.3179, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.3484015475463249e-05, | |
| "loss": 0.3078, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.338220321726736e-05, | |
| "loss": 0.3432, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.3280390959071474e-05, | |
| "loss": 0.267, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.3178578700875585e-05, | |
| "loss": 0.2747, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.30767664426797e-05, | |
| "loss": 0.2626, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.2974954184483813e-05, | |
| "loss": 0.261, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.2873141926287927e-05, | |
| "loss": 0.252, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.8812008054182684, | |
| "eval_loss": 0.30188071727752686, | |
| "eval_runtime": 10.8856, | |
| "eval_samples_per_second": 501.855, | |
| "eval_steps_per_second": 31.418, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.2771329668092038e-05, | |
| "loss": 0.2727, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.2669517409896153e-05, | |
| "loss": 0.2601, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.2567705151700267e-05, | |
| "loss": 0.2622, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 1.2465892893504378e-05, | |
| "loss": 0.272, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.2364080635308493e-05, | |
| "loss": 0.2585, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.2262268377112606e-05, | |
| "loss": 0.2558, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.2160456118916718e-05, | |
| "loss": 0.2606, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.2058643860720831e-05, | |
| "loss": 0.2801, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 1.1956831602524946e-05, | |
| "loss": 0.26, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1855019344329057e-05, | |
| "loss": 0.261, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.8892549881017756, | |
| "eval_loss": 0.27826911211013794, | |
| "eval_runtime": 10.3055, | |
| "eval_samples_per_second": 530.104, | |
| "eval_steps_per_second": 33.186, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.1753207086133171e-05, | |
| "loss": 0.2665, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.1651394827937284e-05, | |
| "loss": 0.2448, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.1549582569741397e-05, | |
| "loss": 0.2665, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.1447770311545512e-05, | |
| "loss": 0.2566, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.1345958053349624e-05, | |
| "loss": 0.2508, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.1244145795153739e-05, | |
| "loss": 0.2411, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.114233353695785e-05, | |
| "loss": 0.2625, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.1040521278761964e-05, | |
| "loss": 0.2557, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.0938709020566076e-05, | |
| "loss": 0.2394, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.083689676237019e-05, | |
| "loss": 0.2718, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_accuracy": 0.8832143510891451, | |
| "eval_loss": 0.287992000579834, | |
| "eval_runtime": 10.6551, | |
| "eval_samples_per_second": 512.712, | |
| "eval_steps_per_second": 32.097, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.0735084504174303e-05, | |
| "loss": 0.2595, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.0633272245978416e-05, | |
| "loss": 0.2532, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.053145998778253e-05, | |
| "loss": 0.2323, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.0429647729586643e-05, | |
| "loss": 0.2586, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.0327835471390758e-05, | |
| "loss": 0.2431, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.0226023213194869e-05, | |
| "loss": 0.2632, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.0124210954998983e-05, | |
| "loss": 0.2674, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.0022398696803094e-05, | |
| "loss": 0.2419, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 9.920586438607209e-06, | |
| "loss": 0.2746, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 9.818774180411322e-06, | |
| "loss": 0.2407, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.8812008054182684, | |
| "eval_loss": 0.30166780948638916, | |
| "eval_runtime": 10.3016, | |
| "eval_samples_per_second": 530.306, | |
| "eval_steps_per_second": 33.199, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 9.716961922215436e-06, | |
| "loss": 0.245, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.615149664019549e-06, | |
| "loss": 0.2627, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 9.513337405823662e-06, | |
| "loss": 0.25, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 9.411525147627775e-06, | |
| "loss": 0.2639, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.309712889431889e-06, | |
| "loss": 0.2454, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.207900631236002e-06, | |
| "loss": 0.2062, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.106088373040115e-06, | |
| "loss": 0.2544, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 9.004276114844227e-06, | |
| "loss": 0.2539, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.902463856648342e-06, | |
| "loss": 0.2587, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.800651598452455e-06, | |
| "loss": 0.254, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.8826652022698152, | |
| "eval_loss": 0.277452290058136, | |
| "eval_runtime": 10.3217, | |
| "eval_samples_per_second": 529.274, | |
| "eval_steps_per_second": 33.134, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 8.698839340256568e-06, | |
| "loss": 0.2558, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 8.59702708206068e-06, | |
| "loss": 0.2455, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 8.495214823864793e-06, | |
| "loss": 0.2428, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 8.393402565668908e-06, | |
| "loss": 0.2499, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 8.29159030747302e-06, | |
| "loss": 0.2456, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 8.189778049277133e-06, | |
| "loss": 0.2348, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 8.087965791081248e-06, | |
| "loss": 0.2496, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 7.98615353288536e-06, | |
| "loss": 0.2589, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 7.884341274689473e-06, | |
| "loss": 0.2524, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 7.782529016493586e-06, | |
| "loss": 0.2611, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.8812008054182684, | |
| "eval_loss": 0.28367963433265686, | |
| "eval_runtime": 10.3638, | |
| "eval_samples_per_second": 527.123, | |
| "eval_steps_per_second": 32.999, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 7.680716758297699e-06, | |
| "loss": 0.2523, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 7.578904500101813e-06, | |
| "loss": 0.2641, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 7.4770922419059255e-06, | |
| "loss": 0.25, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 7.37527998371004e-06, | |
| "loss": 0.2417, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 7.273467725514153e-06, | |
| "loss": 0.2472, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 7.171655467318266e-06, | |
| "loss": 0.2373, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 7.069843209122379e-06, | |
| "loss": 0.2514, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 6.968030950926492e-06, | |
| "loss": 0.2539, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.866218692730605e-06, | |
| "loss": 0.237, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.764406434534719e-06, | |
| "loss": 0.257, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.885227896760022, | |
| "eval_loss": 0.28157150745391846, | |
| "eval_runtime": 10.3073, | |
| "eval_samples_per_second": 530.013, | |
| "eval_steps_per_second": 33.18, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 6.662594176338831e-06, | |
| "loss": 0.2684, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.560781918142944e-06, | |
| "loss": 0.1563, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 6.458969659947059e-06, | |
| "loss": 0.1754, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 6.3571574017511715e-06, | |
| "loss": 0.1781, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 6.255345143555285e-06, | |
| "loss": 0.1645, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 6.153532885359398e-06, | |
| "loss": 0.1746, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 6.051720627163511e-06, | |
| "loss": 0.1903, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 5.949908368967624e-06, | |
| "loss": 0.1872, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 5.848096110771737e-06, | |
| "loss": 0.1929, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 5.74628385257585e-06, | |
| "loss": 0.1645, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.8844956983342486, | |
| "eval_loss": 0.3323143720626831, | |
| "eval_runtime": 10.8137, | |
| "eval_samples_per_second": 505.191, | |
| "eval_steps_per_second": 31.626, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 5.644471594379963e-06, | |
| "loss": 0.1783, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.542659336184077e-06, | |
| "loss": 0.1549, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 5.44084707798819e-06, | |
| "loss": 0.1477, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.339034819792304e-06, | |
| "loss": 0.1756, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 5.237222561596417e-06, | |
| "loss": 0.1844, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 5.1354103034005295e-06, | |
| "loss": 0.1956, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.033598045204643e-06, | |
| "loss": 0.1571, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 4.931785787008756e-06, | |
| "loss": 0.1924, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.82997352881287e-06, | |
| "loss": 0.1459, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.728161270616982e-06, | |
| "loss": 0.1679, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_accuracy": 0.8824821526633718, | |
| "eval_loss": 0.35683655738830566, | |
| "eval_runtime": 10.3196, | |
| "eval_samples_per_second": 529.382, | |
| "eval_steps_per_second": 33.141, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 4.626349012421096e-06, | |
| "loss": 0.1736, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.524536754225209e-06, | |
| "loss": 0.1584, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 4.4227244960293225e-06, | |
| "loss": 0.1891, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 4.320912237833435e-06, | |
| "loss": 0.1603, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.219099979637549e-06, | |
| "loss": 0.1702, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.117287721441662e-06, | |
| "loss": 0.1829, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.015475463245775e-06, | |
| "loss": 0.1668, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.913663205049888e-06, | |
| "loss": 0.173, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 3.8118509468540015e-06, | |
| "loss": 0.1885, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.7100386886581147e-06, | |
| "loss": 0.1643, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_accuracy": 0.8888888888888888, | |
| "eval_loss": 0.32031089067459106, | |
| "eval_runtime": 10.7357, | |
| "eval_samples_per_second": 508.862, | |
| "eval_steps_per_second": 31.856, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3.6082264304622276e-06, | |
| "loss": 0.178, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 3.506414172266341e-06, | |
| "loss": 0.1698, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 3.4046019140704544e-06, | |
| "loss": 0.1762, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.3027896558745677e-06, | |
| "loss": 0.1584, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 3.2009773976786805e-06, | |
| "loss": 0.1684, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.0991651394827937e-06, | |
| "loss": 0.1721, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 2.9973528812869074e-06, | |
| "loss": 0.179, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 2.8955406230910206e-06, | |
| "loss": 0.1654, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.7937283648951334e-06, | |
| "loss": 0.178, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 2.6919161066992467e-06, | |
| "loss": 0.1662, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_accuracy": 0.8877905912502289, | |
| "eval_loss": 0.3240011930465698, | |
| "eval_runtime": 10.3323, | |
| "eval_samples_per_second": 528.728, | |
| "eval_steps_per_second": 33.1, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.5901038485033603e-06, | |
| "loss": 0.1661, | |
| "step": 8550 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 2.4882915903074735e-06, | |
| "loss": 0.1686, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 2.3864793321115864e-06, | |
| "loss": 0.1824, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 2.2846670739156996e-06, | |
| "loss": 0.1845, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.182854815719813e-06, | |
| "loss": 0.1677, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.081042557523926e-06, | |
| "loss": 0.1662, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1.9792302993280393e-06, | |
| "loss": 0.1886, | |
| "step": 8850 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.8774180411321525e-06, | |
| "loss": 0.1631, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 1.7756057829362658e-06, | |
| "loss": 0.16, | |
| "step": 8950 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.6737935247403788e-06, | |
| "loss": 0.1558, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_accuracy": 0.8855939959729087, | |
| "eval_loss": 0.3301653265953064, | |
| "eval_runtime": 10.7134, | |
| "eval_samples_per_second": 509.92, | |
| "eval_steps_per_second": 31.923, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 1.571981266544492e-06, | |
| "loss": 0.1667, | |
| "step": 9050 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.4701690083486052e-06, | |
| "loss": 0.1896, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.3683567501527185e-06, | |
| "loss": 0.1573, | |
| "step": 9150 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.2665444919568317e-06, | |
| "loss": 0.1582, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.164732233760945e-06, | |
| "loss": 0.1536, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.0629199755650582e-06, | |
| "loss": 0.1747, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 9.611077173691712e-07, | |
| "loss": 0.1864, | |
| "step": 9350 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 8.592954591732845e-07, | |
| "loss": 0.1613, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.574832009773978e-07, | |
| "loss": 0.1771, | |
| "step": 9450 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 6.55670942781511e-07, | |
| "loss": 0.1614, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.8872414424308988, | |
| "eval_loss": 0.32990914583206177, | |
| "eval_runtime": 10.346, | |
| "eval_samples_per_second": 528.031, | |
| "eval_steps_per_second": 33.056, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.538586845856241e-07, | |
| "loss": 0.1729, | |
| "step": 9550 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.5204642638973736e-07, | |
| "loss": 0.1546, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.502341681938506e-07, | |
| "loss": 0.1551, | |
| "step": 9650 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 2.484219099979638e-07, | |
| "loss": 0.1717, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 1.4660965180207698e-07, | |
| "loss": 0.1523, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.4797393606190183e-08, | |
| "loss": 0.1537, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 9822, | |
| "total_flos": 2.08897510597413e+16, | |
| "train_loss": 0.2648926304110537, | |
| "train_runtime": 2482.4634, | |
| "train_samples_per_second": 126.58, | |
| "train_steps_per_second": 3.957 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 9822, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.08897510597413e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |