| { | |
| "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": 0.5946, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.9796375483608227e-05, | |
| "loss": 0.4843, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.9694563225412342e-05, | |
| "loss": 0.482, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.9592750967216453e-05, | |
| "loss": 0.4304, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.9490938709020567e-05, | |
| "loss": 0.434, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.938912645082468e-05, | |
| "loss": 0.4322, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.9287314192628793e-05, | |
| "loss": 0.4277, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.9185501934432908e-05, | |
| "loss": 0.4044, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9083689676237022e-05, | |
| "loss": 0.3945, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.8981877418041133e-05, | |
| "loss": 0.3938, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.8495332235035694, | |
| "eval_loss": 0.34940680861473083, | |
| "eval_runtime": 10.3017, | |
| "eval_samples_per_second": 530.299, | |
| "eval_steps_per_second": 33.198, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.8880065159845248e-05, | |
| "loss": 0.3993, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.8778252901649362e-05, | |
| "loss": 0.4018, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.8676440643453473e-05, | |
| "loss": 0.4092, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.8574628385257588e-05, | |
| "loss": 0.4052, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.84728161270617e-05, | |
| "loss": 0.3735, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.8371003868865813e-05, | |
| "loss": 0.4142, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.8269191610669924e-05, | |
| "loss": 0.4004, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.816737935247404e-05, | |
| "loss": 0.3857, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.8065567094278153e-05, | |
| "loss": 0.3771, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.7963754836082265e-05, | |
| "loss": 0.3712, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.8570382573677466, | |
| "eval_loss": 0.3266446590423584, | |
| "eval_runtime": 10.3789, | |
| "eval_samples_per_second": 526.358, | |
| "eval_steps_per_second": 32.952, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.786194257788638e-05, | |
| "loss": 0.365, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.776013031969049e-05, | |
| "loss": 0.3537, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1.7658318061494605e-05, | |
| "loss": 0.359, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.755650580329872e-05, | |
| "loss": 0.3704, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.7454693545102834e-05, | |
| "loss": 0.3598, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 1.7352881286906945e-05, | |
| "loss": 0.3571, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.725106902871106e-05, | |
| "loss": 0.3599, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.7149256770515174e-05, | |
| "loss": 0.3509, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.7047444512319285e-05, | |
| "loss": 0.3563, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.69456322541234e-05, | |
| "loss": 0.3837, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_accuracy": 0.8654585392641406, | |
| "eval_loss": 0.31736862659454346, | |
| "eval_runtime": 10.379, | |
| "eval_samples_per_second": 526.349, | |
| "eval_steps_per_second": 32.951, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.684381999592751e-05, | |
| "loss": 0.3574, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.6742007737731625e-05, | |
| "loss": 0.3587, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.6640195479535736e-05, | |
| "loss": 0.3575, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.653838322133985e-05, | |
| "loss": 0.3593, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1.6436570963143962e-05, | |
| "loss": 0.3429, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.6334758704948076e-05, | |
| "loss": 0.3663, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.623294644675219e-05, | |
| "loss": 0.3661, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.6131134188556302e-05, | |
| "loss": 0.3725, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.6029321930360416e-05, | |
| "loss": 0.3573, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.592750967216453e-05, | |
| "loss": 0.3466, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.8784550613216182, | |
| "eval_loss": 0.2956714928150177, | |
| "eval_runtime": 10.4846, | |
| "eval_samples_per_second": 521.05, | |
| "eval_steps_per_second": 32.619, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.5825697413968642e-05, | |
| "loss": 0.35, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.5723885155772757e-05, | |
| "loss": 0.334, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.562207289757687e-05, | |
| "loss": 0.3352, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.5520260639380982e-05, | |
| "loss": 0.3487, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.5418448381185097e-05, | |
| "loss": 0.3272, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.531663612298921e-05, | |
| "loss": 0.3314, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.5214823864793322e-05, | |
| "loss": 0.3415, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.5113011606597437e-05, | |
| "loss": 0.3638, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.5011199348401548e-05, | |
| "loss": 0.35, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.4909387090205662e-05, | |
| "loss": 0.3084, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.871499176276771, | |
| "eval_loss": 0.30931228399276733, | |
| "eval_runtime": 10.4187, | |
| "eval_samples_per_second": 524.348, | |
| "eval_steps_per_second": 32.826, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.4807574832009775e-05, | |
| "loss": 0.3414, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.4705762573813888e-05, | |
| "loss": 0.3368, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.4603950315618001e-05, | |
| "loss": 0.334, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.4502138057422115e-05, | |
| "loss": 0.3116, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.440032579922623e-05, | |
| "loss": 0.3253, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.4298513541030341e-05, | |
| "loss": 0.3395, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.4196701282834456e-05, | |
| "loss": 0.3159, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.4094889024638567e-05, | |
| "loss": 0.3633, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.3993076766442681e-05, | |
| "loss": 0.3121, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.3891264508246794e-05, | |
| "loss": 0.322, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.8731466227347611, | |
| "eval_loss": 0.2949598729610443, | |
| "eval_runtime": 10.4266, | |
| "eval_samples_per_second": 523.949, | |
| "eval_steps_per_second": 32.801, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.3789452250050907e-05, | |
| "loss": 0.3306, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.368763999185502e-05, | |
| "loss": 0.3367, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.3585827733659134e-05, | |
| "loss": 0.3183, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.3484015475463249e-05, | |
| "loss": 0.3081, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.338220321726736e-05, | |
| "loss": 0.3566, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.3280390959071474e-05, | |
| "loss": 0.2855, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.3178578700875585e-05, | |
| "loss": 0.2711, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.30767664426797e-05, | |
| "loss": 0.2663, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.2974954184483813e-05, | |
| "loss": 0.265, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.2873141926287927e-05, | |
| "loss": 0.273, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.8833974006955885, | |
| "eval_loss": 0.28723686933517456, | |
| "eval_runtime": 10.3846, | |
| "eval_samples_per_second": 526.066, | |
| "eval_steps_per_second": 32.933, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.2771329668092038e-05, | |
| "loss": 0.2624, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.2669517409896153e-05, | |
| "loss": 0.2681, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.2567705151700267e-05, | |
| "loss": 0.269, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 1.2465892893504378e-05, | |
| "loss": 0.2604, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.2364080635308493e-05, | |
| "loss": 0.2716, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.2262268377112606e-05, | |
| "loss": 0.2639, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.2160456118916718e-05, | |
| "loss": 0.2433, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.2058643860720831e-05, | |
| "loss": 0.2593, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 1.1956831602524946e-05, | |
| "loss": 0.2519, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1855019344329057e-05, | |
| "loss": 0.2628, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.8793703093538349, | |
| "eval_loss": 0.3109899163246155, | |
| "eval_runtime": 10.3922, | |
| "eval_samples_per_second": 525.685, | |
| "eval_steps_per_second": 32.909, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.1753207086133171e-05, | |
| "loss": 0.2555, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.1651394827937284e-05, | |
| "loss": 0.2585, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.1549582569741397e-05, | |
| "loss": 0.24, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.1447770311545512e-05, | |
| "loss": 0.2556, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.1345958053349624e-05, | |
| "loss": 0.2417, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.1244145795153739e-05, | |
| "loss": 0.2347, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.114233353695785e-05, | |
| "loss": 0.256, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.1040521278761964e-05, | |
| "loss": 0.2657, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.0938709020566076e-05, | |
| "loss": 0.2668, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.083689676237019e-05, | |
| "loss": 0.2732, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_accuracy": 0.8797364085667215, | |
| "eval_loss": 0.29097074270248413, | |
| "eval_runtime": 10.3775, | |
| "eval_samples_per_second": 526.429, | |
| "eval_steps_per_second": 32.956, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.0735084504174303e-05, | |
| "loss": 0.2339, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.0633272245978416e-05, | |
| "loss": 0.2486, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.053145998778253e-05, | |
| "loss": 0.2472, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.0429647729586643e-05, | |
| "loss": 0.2575, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.0327835471390758e-05, | |
| "loss": 0.2502, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.0226023213194869e-05, | |
| "loss": 0.2685, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.0124210954998983e-05, | |
| "loss": 0.2568, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.0022398696803094e-05, | |
| "loss": 0.2522, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 9.920586438607209e-06, | |
| "loss": 0.2504, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 9.818774180411322e-06, | |
| "loss": 0.2592, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.8848617975471352, | |
| "eval_loss": 0.2854502201080322, | |
| "eval_runtime": 10.4181, | |
| "eval_samples_per_second": 524.375, | |
| "eval_steps_per_second": 32.827, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 9.716961922215436e-06, | |
| "loss": 0.2663, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.615149664019549e-06, | |
| "loss": 0.2564, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 9.513337405823662e-06, | |
| "loss": 0.2551, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 9.411525147627775e-06, | |
| "loss": 0.2796, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.309712889431889e-06, | |
| "loss": 0.2661, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.207900631236002e-06, | |
| "loss": 0.2592, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.106088373040115e-06, | |
| "loss": 0.229, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 9.004276114844227e-06, | |
| "loss": 0.2454, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.902463856648342e-06, | |
| "loss": 0.2412, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.800651598452455e-06, | |
| "loss": 0.241, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.8861431447922387, | |
| "eval_loss": 0.297364205121994, | |
| "eval_runtime": 10.4358, | |
| "eval_samples_per_second": 523.486, | |
| "eval_steps_per_second": 32.772, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 8.698839340256568e-06, | |
| "loss": 0.2584, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 8.59702708206068e-06, | |
| "loss": 0.2293, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 8.495214823864793e-06, | |
| "loss": 0.2464, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 8.393402565668908e-06, | |
| "loss": 0.2548, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 8.29159030747302e-06, | |
| "loss": 0.2152, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 8.189778049277133e-06, | |
| "loss": 0.2499, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 8.087965791081248e-06, | |
| "loss": 0.2472, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 7.98615353288536e-06, | |
| "loss": 0.2411, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 7.884341274689473e-06, | |
| "loss": 0.2372, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 7.782529016493586e-06, | |
| "loss": 0.2256, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.8850448471535787, | |
| "eval_loss": 0.29136955738067627, | |
| "eval_runtime": 10.3944, | |
| "eval_samples_per_second": 525.569, | |
| "eval_steps_per_second": 32.902, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 7.680716758297699e-06, | |
| "loss": 0.2335, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 7.578904500101813e-06, | |
| "loss": 0.2556, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 7.4770922419059255e-06, | |
| "loss": 0.2491, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 7.37527998371004e-06, | |
| "loss": 0.2381, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 7.273467725514153e-06, | |
| "loss": 0.2374, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 7.171655467318266e-06, | |
| "loss": 0.2598, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 7.069843209122379e-06, | |
| "loss": 0.2423, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 6.968030950926492e-06, | |
| "loss": 0.2393, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.866218692730605e-06, | |
| "loss": 0.2402, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.764406434534719e-06, | |
| "loss": 0.2402, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.8883397400695588, | |
| "eval_loss": 0.27589744329452515, | |
| "eval_runtime": 10.4485, | |
| "eval_samples_per_second": 522.851, | |
| "eval_steps_per_second": 32.732, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 6.662594176338831e-06, | |
| "loss": 0.2396, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.560781918142944e-06, | |
| "loss": 0.1916, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 6.458969659947059e-06, | |
| "loss": 0.1656, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 6.3571574017511715e-06, | |
| "loss": 0.1705, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 6.255345143555285e-06, | |
| "loss": 0.1717, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 6.153532885359398e-06, | |
| "loss": 0.1718, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 6.051720627163511e-06, | |
| "loss": 0.1676, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 5.949908368967624e-06, | |
| "loss": 0.1542, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 5.848096110771737e-06, | |
| "loss": 0.176, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 5.74628385257585e-06, | |
| "loss": 0.1958, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.8879736408566722, | |
| "eval_loss": 0.30796074867248535, | |
| "eval_runtime": 10.4169, | |
| "eval_samples_per_second": 524.435, | |
| "eval_steps_per_second": 32.831, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 5.644471594379963e-06, | |
| "loss": 0.1696, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.542659336184077e-06, | |
| "loss": 0.1586, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 5.44084707798819e-06, | |
| "loss": 0.1766, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.339034819792304e-06, | |
| "loss": 0.1646, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 5.237222561596417e-06, | |
| "loss": 0.1738, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 5.1354103034005295e-06, | |
| "loss": 0.1686, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.033598045204643e-06, | |
| "loss": 0.1735, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 4.931785787008756e-06, | |
| "loss": 0.1662, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.82997352881287e-06, | |
| "loss": 0.1683, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.728161270616982e-06, | |
| "loss": 0.1684, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_accuracy": 0.8846787479406919, | |
| "eval_loss": 0.31903892755508423, | |
| "eval_runtime": 10.398, | |
| "eval_samples_per_second": 525.391, | |
| "eval_steps_per_second": 32.891, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 4.626349012421096e-06, | |
| "loss": 0.1694, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.524536754225209e-06, | |
| "loss": 0.1708, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 4.4227244960293225e-06, | |
| "loss": 0.1652, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 4.320912237833435e-06, | |
| "loss": 0.1873, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.219099979637549e-06, | |
| "loss": 0.1507, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.117287721441662e-06, | |
| "loss": 0.1812, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.015475463245775e-06, | |
| "loss": 0.1656, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.913663205049888e-06, | |
| "loss": 0.163, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 3.8118509468540015e-06, | |
| "loss": 0.168, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.7100386886581147e-06, | |
| "loss": 0.1472, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_accuracy": 0.8870583928244554, | |
| "eval_loss": 0.33048346638679504, | |
| "eval_runtime": 10.5266, | |
| "eval_samples_per_second": 518.973, | |
| "eval_steps_per_second": 32.489, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3.6082264304622276e-06, | |
| "loss": 0.1606, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 3.506414172266341e-06, | |
| "loss": 0.1596, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 3.4046019140704544e-06, | |
| "loss": 0.1809, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.3027896558745677e-06, | |
| "loss": 0.178, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 3.2009773976786805e-06, | |
| "loss": 0.1612, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.0991651394827937e-06, | |
| "loss": 0.1895, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 2.9973528812869074e-06, | |
| "loss": 0.1818, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 2.8955406230910206e-06, | |
| "loss": 0.1757, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.7937283648951334e-06, | |
| "loss": 0.1616, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 2.6919161066992467e-06, | |
| "loss": 0.1601, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_accuracy": 0.8835804503020318, | |
| "eval_loss": 0.3297683894634247, | |
| "eval_runtime": 10.3814, | |
| "eval_samples_per_second": 526.23, | |
| "eval_steps_per_second": 32.944, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.5901038485033603e-06, | |
| "loss": 0.148, | |
| "step": 8550 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 2.4882915903074735e-06, | |
| "loss": 0.1943, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 2.3864793321115864e-06, | |
| "loss": 0.1779, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 2.2846670739156996e-06, | |
| "loss": 0.1596, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.182854815719813e-06, | |
| "loss": 0.1552, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.081042557523926e-06, | |
| "loss": 0.1614, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1.9792302993280393e-06, | |
| "loss": 0.1897, | |
| "step": 8850 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.8774180411321525e-06, | |
| "loss": 0.158, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 1.7756057829362658e-06, | |
| "loss": 0.1484, | |
| "step": 8950 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.6737935247403788e-06, | |
| "loss": 0.1857, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_accuracy": 0.8846787479406919, | |
| "eval_loss": 0.32735589146614075, | |
| "eval_runtime": 10.3329, | |
| "eval_samples_per_second": 528.702, | |
| "eval_steps_per_second": 33.098, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 1.571981266544492e-06, | |
| "loss": 0.1575, | |
| "step": 9050 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.4701690083486052e-06, | |
| "loss": 0.1565, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.3683567501527185e-06, | |
| "loss": 0.1821, | |
| "step": 9150 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.2665444919568317e-06, | |
| "loss": 0.1793, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.164732233760945e-06, | |
| "loss": 0.1502, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.0629199755650582e-06, | |
| "loss": 0.1395, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 9.611077173691712e-07, | |
| "loss": 0.1574, | |
| "step": 9350 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 8.592954591732845e-07, | |
| "loss": 0.1724, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.574832009773978e-07, | |
| "loss": 0.1759, | |
| "step": 9450 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 6.55670942781511e-07, | |
| "loss": 0.1667, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.8841295991213619, | |
| "eval_loss": 0.3255881667137146, | |
| "eval_runtime": 10.3472, | |
| "eval_samples_per_second": 527.97, | |
| "eval_steps_per_second": 33.052, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.538586845856241e-07, | |
| "loss": 0.1503, | |
| "step": 9550 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.5204642638973736e-07, | |
| "loss": 0.1537, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.502341681938506e-07, | |
| "loss": 0.1878, | |
| "step": 9650 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 2.484219099979638e-07, | |
| "loss": 0.1718, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 1.4660965180207698e-07, | |
| "loss": 0.1545, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.4797393606190183e-08, | |
| "loss": 0.1533, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 9822, | |
| "total_flos": 2.086631721748452e+16, | |
| "train_loss": 0.2627993202384245, | |
| "train_runtime": 2379.8057, | |
| "train_samples_per_second": 132.04, | |
| "train_steps_per_second": 4.127 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 9822, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.086631721748452e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |