| { | |
| "best_metric": 0.38806891441345215, | |
| "best_model_checkpoint": "./model_fine-tune/glot/xlm-r/ckb-Arab/checkpoint-95500", | |
| "epoch": 173.32123411978222, | |
| "eval_steps": 500, | |
| "global_step": 95500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9074410163339383, | |
| "grad_norm": 3.4398534297943115, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.6403, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9074410163339383, | |
| "eval_accuracy": 0.752505506780714, | |
| "eval_loss": 1.153350591659546, | |
| "eval_runtime": 53.4289, | |
| "eval_samples_per_second": 82.446, | |
| "eval_steps_per_second": 2.583, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8148820326678767, | |
| "grad_norm": 2.854556083679199, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.1266, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.8148820326678767, | |
| "eval_accuracy": 0.7904873277235048, | |
| "eval_loss": 0.9590327143669128, | |
| "eval_runtime": 53.4057, | |
| "eval_samples_per_second": 82.482, | |
| "eval_steps_per_second": 2.584, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.722323049001815, | |
| "grad_norm": 2.9346060752868652, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 0.9785, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.722323049001815, | |
| "eval_accuracy": 0.8084301658954617, | |
| "eval_loss": 0.8642853498458862, | |
| "eval_runtime": 59.5918, | |
| "eval_samples_per_second": 73.92, | |
| "eval_steps_per_second": 2.316, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.629764065335753, | |
| "grad_norm": 2.695502996444702, | |
| "learning_rate": 9.8e-05, | |
| "loss": 0.8933, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.629764065335753, | |
| "eval_accuracy": 0.8202711962432342, | |
| "eval_loss": 0.8070259094238281, | |
| "eval_runtime": 54.3445, | |
| "eval_samples_per_second": 81.057, | |
| "eval_steps_per_second": 2.539, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.537205081669692, | |
| "grad_norm": 2.3147311210632324, | |
| "learning_rate": 9.75e-05, | |
| "loss": 0.8354, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.537205081669692, | |
| "eval_accuracy": 0.8293452619955832, | |
| "eval_loss": 0.7662876844406128, | |
| "eval_runtime": 54.1554, | |
| "eval_samples_per_second": 81.34, | |
| "eval_steps_per_second": 2.548, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.44464609800363, | |
| "grad_norm": 2.6086342334747314, | |
| "learning_rate": 9.7e-05, | |
| "loss": 0.7909, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.44464609800363, | |
| "eval_accuracy": 0.8355002523340903, | |
| "eval_loss": 0.7356261610984802, | |
| "eval_runtime": 53.6095, | |
| "eval_samples_per_second": 82.168, | |
| "eval_steps_per_second": 2.574, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.352087114337568, | |
| "grad_norm": 2.1992199420928955, | |
| "learning_rate": 9.65e-05, | |
| "loss": 0.7546, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.352087114337568, | |
| "eval_accuracy": 0.8423681886232234, | |
| "eval_loss": 0.6986920237541199, | |
| "eval_runtime": 55.6018, | |
| "eval_samples_per_second": 79.224, | |
| "eval_steps_per_second": 2.482, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.259528130671506, | |
| "grad_norm": 2.3473734855651855, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.7292, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.259528130671506, | |
| "eval_accuracy": 0.8470036500248151, | |
| "eval_loss": 0.6758582592010498, | |
| "eval_runtime": 54.4017, | |
| "eval_samples_per_second": 80.972, | |
| "eval_steps_per_second": 2.537, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.166969147005444, | |
| "grad_norm": 2.585818290710449, | |
| "learning_rate": 9.55e-05, | |
| "loss": 0.7025, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.166969147005444, | |
| "eval_accuracy": 0.8527411894826806, | |
| "eval_loss": 0.6568289399147034, | |
| "eval_runtime": 57.7681, | |
| "eval_samples_per_second": 76.253, | |
| "eval_steps_per_second": 2.389, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.074410163339383, | |
| "grad_norm": 2.4189865589141846, | |
| "learning_rate": 9.5e-05, | |
| "loss": 0.68, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.074410163339383, | |
| "eval_accuracy": 0.8559500862970382, | |
| "eval_loss": 0.6354258060455322, | |
| "eval_runtime": 59.1091, | |
| "eval_samples_per_second": 74.523, | |
| "eval_steps_per_second": 2.335, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.98185117967332, | |
| "grad_norm": 2.443392515182495, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 0.6645, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.98185117967332, | |
| "eval_accuracy": 0.8591041578043516, | |
| "eval_loss": 0.6224693655967712, | |
| "eval_runtime": 53.9006, | |
| "eval_samples_per_second": 81.725, | |
| "eval_steps_per_second": 2.56, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.88929219600726, | |
| "grad_norm": 2.1055374145507812, | |
| "learning_rate": 9.4e-05, | |
| "loss": 0.6385, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.88929219600726, | |
| "eval_accuracy": 0.8609373044675259, | |
| "eval_loss": 0.6164940595626831, | |
| "eval_runtime": 53.8382, | |
| "eval_samples_per_second": 81.819, | |
| "eval_steps_per_second": 2.563, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.796733212341199, | |
| "grad_norm": 1.9976396560668945, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 0.6305, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 11.796733212341199, | |
| "eval_accuracy": 0.8621495473876636, | |
| "eval_loss": 0.6023448705673218, | |
| "eval_runtime": 54.3137, | |
| "eval_samples_per_second": 81.103, | |
| "eval_steps_per_second": 2.541, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.704174228675136, | |
| "grad_norm": 2.6231155395507812, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 0.6143, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 12.704174228675136, | |
| "eval_accuracy": 0.8642217993678152, | |
| "eval_loss": 0.5955979824066162, | |
| "eval_runtime": 55.8185, | |
| "eval_samples_per_second": 78.916, | |
| "eval_steps_per_second": 2.472, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.611615245009075, | |
| "grad_norm": 2.0853583812713623, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 0.6012, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 13.611615245009075, | |
| "eval_accuracy": 0.8669231710723286, | |
| "eval_loss": 0.5834583044052124, | |
| "eval_runtime": 52.8539, | |
| "eval_samples_per_second": 83.343, | |
| "eval_steps_per_second": 2.611, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.519056261343012, | |
| "grad_norm": 1.9873307943344116, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.59, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 14.519056261343012, | |
| "eval_accuracy": 0.8700438033381165, | |
| "eval_loss": 0.5704456567764282, | |
| "eval_runtime": 52.8195, | |
| "eval_samples_per_second": 83.397, | |
| "eval_steps_per_second": 2.613, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.426497277676951, | |
| "grad_norm": 2.106224536895752, | |
| "learning_rate": 9.15e-05, | |
| "loss": 0.5781, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 15.426497277676951, | |
| "eval_accuracy": 0.8698398656877739, | |
| "eval_loss": 0.5724136233329773, | |
| "eval_runtime": 59.543, | |
| "eval_samples_per_second": 73.98, | |
| "eval_steps_per_second": 2.318, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.33393829401089, | |
| "grad_norm": 2.0043249130249023, | |
| "learning_rate": 9.1e-05, | |
| "loss": 0.5675, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 16.33393829401089, | |
| "eval_accuracy": 0.8715064286118855, | |
| "eval_loss": 0.5624237656593323, | |
| "eval_runtime": 53.5321, | |
| "eval_samples_per_second": 82.287, | |
| "eval_steps_per_second": 2.578, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.24137931034483, | |
| "grad_norm": 2.078568935394287, | |
| "learning_rate": 9.05e-05, | |
| "loss": 0.5566, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 17.24137931034483, | |
| "eval_accuracy": 0.872356935014549, | |
| "eval_loss": 0.5570796728134155, | |
| "eval_runtime": 52.8375, | |
| "eval_samples_per_second": 83.369, | |
| "eval_steps_per_second": 2.612, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.148820326678766, | |
| "grad_norm": 2.026803731918335, | |
| "learning_rate": 9e-05, | |
| "loss": 0.5533, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 18.148820326678766, | |
| "eval_accuracy": 0.875522907428112, | |
| "eval_loss": 0.5457433462142944, | |
| "eval_runtime": 52.7025, | |
| "eval_samples_per_second": 83.582, | |
| "eval_steps_per_second": 2.618, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.056261343012704, | |
| "grad_norm": 1.8563569784164429, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 0.5396, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 19.056261343012704, | |
| "eval_accuracy": 0.8741645771312478, | |
| "eval_loss": 0.5479493737220764, | |
| "eval_runtime": 52.8826, | |
| "eval_samples_per_second": 83.298, | |
| "eval_steps_per_second": 2.61, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 19.96370235934664, | |
| "grad_norm": 2.1099376678466797, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 0.5359, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 19.96370235934664, | |
| "eval_accuracy": 0.8773543802678719, | |
| "eval_loss": 0.5376391410827637, | |
| "eval_runtime": 53.3193, | |
| "eval_samples_per_second": 82.615, | |
| "eval_steps_per_second": 2.588, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 20.87114337568058, | |
| "grad_norm": 2.5443553924560547, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 0.5246, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 20.87114337568058, | |
| "eval_accuracy": 0.8773027699144483, | |
| "eval_loss": 0.5346177816390991, | |
| "eval_runtime": 51.9467, | |
| "eval_samples_per_second": 84.799, | |
| "eval_steps_per_second": 2.657, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 21.77858439201452, | |
| "grad_norm": 2.047163963317871, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.5173, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 21.77858439201452, | |
| "eval_accuracy": 0.880544279055781, | |
| "eval_loss": 0.5207871794700623, | |
| "eval_runtime": 62.4765, | |
| "eval_samples_per_second": 70.506, | |
| "eval_steps_per_second": 2.209, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 22.686025408348456, | |
| "grad_norm": 1.9300446510314941, | |
| "learning_rate": 8.75e-05, | |
| "loss": 0.5111, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 22.686025408348456, | |
| "eval_accuracy": 0.8803666290669171, | |
| "eval_loss": 0.5259021520614624, | |
| "eval_runtime": 53.0987, | |
| "eval_samples_per_second": 82.959, | |
| "eval_steps_per_second": 2.599, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 23.593466424682397, | |
| "grad_norm": 2.314628839492798, | |
| "learning_rate": 8.7e-05, | |
| "loss": 0.505, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 23.593466424682397, | |
| "eval_accuracy": 0.8821161209658513, | |
| "eval_loss": 0.5170288681983948, | |
| "eval_runtime": 60.4353, | |
| "eval_samples_per_second": 72.888, | |
| "eval_steps_per_second": 2.283, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 24.500907441016334, | |
| "grad_norm": 2.187793493270874, | |
| "learning_rate": 8.65e-05, | |
| "loss": 0.4995, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 24.500907441016334, | |
| "eval_accuracy": 0.8803155533947083, | |
| "eval_loss": 0.5203161835670471, | |
| "eval_runtime": 54.2267, | |
| "eval_samples_per_second": 81.233, | |
| "eval_steps_per_second": 2.545, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 25.40834845735027, | |
| "grad_norm": 1.8629887104034424, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.4933, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 25.40834845735027, | |
| "eval_accuracy": 0.8827071990702727, | |
| "eval_loss": 0.5080223679542542, | |
| "eval_runtime": 53.122, | |
| "eval_samples_per_second": 82.922, | |
| "eval_steps_per_second": 2.598, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 26.31578947368421, | |
| "grad_norm": 1.95268976688385, | |
| "learning_rate": 8.55e-05, | |
| "loss": 0.4841, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 26.31578947368421, | |
| "eval_accuracy": 0.8852798720942991, | |
| "eval_loss": 0.4982340335845947, | |
| "eval_runtime": 55.7058, | |
| "eval_samples_per_second": 79.076, | |
| "eval_steps_per_second": 2.477, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 27.22323049001815, | |
| "grad_norm": 1.9664937257766724, | |
| "learning_rate": 8.5e-05, | |
| "loss": 0.4769, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 27.22323049001815, | |
| "eval_accuracy": 0.8844890070822018, | |
| "eval_loss": 0.5071918964385986, | |
| "eval_runtime": 57.0471, | |
| "eval_samples_per_second": 77.217, | |
| "eval_steps_per_second": 2.419, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 28.130671506352087, | |
| "grad_norm": 2.0566840171813965, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 0.4782, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 28.130671506352087, | |
| "eval_accuracy": 0.885741458637098, | |
| "eval_loss": 0.4975322186946869, | |
| "eval_runtime": 53.7809, | |
| "eval_samples_per_second": 81.906, | |
| "eval_steps_per_second": 2.566, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 29.038112522686024, | |
| "grad_norm": 1.969655990600586, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.4728, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 29.038112522686024, | |
| "eval_accuracy": 0.8861595126788497, | |
| "eval_loss": 0.498710572719574, | |
| "eval_runtime": 58.2764, | |
| "eval_samples_per_second": 75.588, | |
| "eval_steps_per_second": 2.368, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 29.945553539019965, | |
| "grad_norm": 1.9814519882202148, | |
| "learning_rate": 8.35e-05, | |
| "loss": 0.4648, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 29.945553539019965, | |
| "eval_accuracy": 0.8870825900148245, | |
| "eval_loss": 0.49478381872177124, | |
| "eval_runtime": 53.8125, | |
| "eval_samples_per_second": 81.858, | |
| "eval_steps_per_second": 2.564, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 30.852994555353902, | |
| "grad_norm": 1.8299716711044312, | |
| "learning_rate": 8.3e-05, | |
| "loss": 0.4597, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 30.852994555353902, | |
| "eval_accuracy": 0.8868082072258864, | |
| "eval_loss": 0.49765515327453613, | |
| "eval_runtime": 54.5146, | |
| "eval_samples_per_second": 80.804, | |
| "eval_steps_per_second": 2.531, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 31.76043557168784, | |
| "grad_norm": 2.192680597305298, | |
| "learning_rate": 8.25e-05, | |
| "loss": 0.4569, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 31.76043557168784, | |
| "eval_accuracy": 0.8874967002300411, | |
| "eval_loss": 0.49416524171829224, | |
| "eval_runtime": 53.3923, | |
| "eval_samples_per_second": 82.503, | |
| "eval_steps_per_second": 2.585, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 32.66787658802178, | |
| "grad_norm": 2.5133163928985596, | |
| "learning_rate": 8.2e-05, | |
| "loss": 0.4488, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 32.66787658802178, | |
| "eval_accuracy": 0.8887065044419662, | |
| "eval_loss": 0.49053409695625305, | |
| "eval_runtime": 56.0002, | |
| "eval_samples_per_second": 78.66, | |
| "eval_steps_per_second": 2.464, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 33.57531760435572, | |
| "grad_norm": 1.596177339553833, | |
| "learning_rate": 8.15e-05, | |
| "loss": 0.4456, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 33.57531760435572, | |
| "eval_accuracy": 0.8882017291247682, | |
| "eval_loss": 0.4841141700744629, | |
| "eval_runtime": 53.2223, | |
| "eval_samples_per_second": 82.766, | |
| "eval_steps_per_second": 2.593, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 34.48275862068966, | |
| "grad_norm": 2.740516185760498, | |
| "learning_rate": 8.1e-05, | |
| "loss": 0.4439, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 34.48275862068966, | |
| "eval_accuracy": 0.8906933029564148, | |
| "eval_loss": 0.473172128200531, | |
| "eval_runtime": 53.5999, | |
| "eval_samples_per_second": 82.183, | |
| "eval_steps_per_second": 2.575, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 35.39019963702359, | |
| "grad_norm": 1.7900762557983398, | |
| "learning_rate": 8.05e-05, | |
| "loss": 0.435, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 35.39019963702359, | |
| "eval_accuracy": 0.8905509902844176, | |
| "eval_loss": 0.4774630069732666, | |
| "eval_runtime": 53.7773, | |
| "eval_samples_per_second": 81.912, | |
| "eval_steps_per_second": 2.566, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 36.29764065335753, | |
| "grad_norm": 1.9263832569122314, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4355, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 36.29764065335753, | |
| "eval_accuracy": 0.889703235016953, | |
| "eval_loss": 0.4819239377975464, | |
| "eval_runtime": 54.1016, | |
| "eval_samples_per_second": 81.421, | |
| "eval_steps_per_second": 2.551, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 37.20508166969147, | |
| "grad_norm": 1.9390649795532227, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 0.4327, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 37.20508166969147, | |
| "eval_accuracy": 0.8891170881908429, | |
| "eval_loss": 0.48580387234687805, | |
| "eval_runtime": 59.5973, | |
| "eval_samples_per_second": 73.913, | |
| "eval_steps_per_second": 2.316, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 38.11252268602541, | |
| "grad_norm": 1.9884870052337646, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 0.4254, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 38.11252268602541, | |
| "eval_accuracy": 0.8915290748428344, | |
| "eval_loss": 0.4742184281349182, | |
| "eval_runtime": 54.3464, | |
| "eval_samples_per_second": 81.054, | |
| "eval_steps_per_second": 2.539, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 39.01996370235935, | |
| "grad_norm": 1.5046188831329346, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 0.4229, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 39.01996370235935, | |
| "eval_accuracy": 0.8925229541919623, | |
| "eval_loss": 0.47058817744255066, | |
| "eval_runtime": 56.2804, | |
| "eval_samples_per_second": 78.269, | |
| "eval_steps_per_second": 2.452, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 39.92740471869328, | |
| "grad_norm": 1.9617971181869507, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.4174, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 39.92740471869328, | |
| "eval_accuracy": 0.8919159314703821, | |
| "eval_loss": 0.4736374616622925, | |
| "eval_runtime": 53.1824, | |
| "eval_samples_per_second": 82.828, | |
| "eval_steps_per_second": 2.595, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 40.83484573502722, | |
| "grad_norm": 1.7383509874343872, | |
| "learning_rate": 7.75e-05, | |
| "loss": 0.4151, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 40.83484573502722, | |
| "eval_accuracy": 0.8929338481208785, | |
| "eval_loss": 0.46910572052001953, | |
| "eval_runtime": 53.0493, | |
| "eval_samples_per_second": 83.036, | |
| "eval_steps_per_second": 2.601, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 41.74228675136116, | |
| "grad_norm": 1.895717740058899, | |
| "learning_rate": 7.7e-05, | |
| "loss": 0.4137, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 41.74228675136116, | |
| "eval_accuracy": 0.8939702847059194, | |
| "eval_loss": 0.46362048387527466, | |
| "eval_runtime": 53.7274, | |
| "eval_samples_per_second": 81.988, | |
| "eval_steps_per_second": 2.569, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 42.6497277676951, | |
| "grad_norm": 1.9185525178909302, | |
| "learning_rate": 7.65e-05, | |
| "loss": 0.4124, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 42.6497277676951, | |
| "eval_accuracy": 0.8955085956295108, | |
| "eval_loss": 0.4603004455566406, | |
| "eval_runtime": 52.7296, | |
| "eval_samples_per_second": 83.539, | |
| "eval_steps_per_second": 2.617, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 43.55716878402904, | |
| "grad_norm": 1.9594053030014038, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.4022, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 43.55716878402904, | |
| "eval_accuracy": 0.894325563921544, | |
| "eval_loss": 0.4673307240009308, | |
| "eval_runtime": 52.947, | |
| "eval_samples_per_second": 83.196, | |
| "eval_steps_per_second": 2.606, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 44.46460980036298, | |
| "grad_norm": 1.761846899986267, | |
| "learning_rate": 7.55e-05, | |
| "loss": 0.4035, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 44.46460980036298, | |
| "eval_accuracy": 0.8957788122798188, | |
| "eval_loss": 0.45509716868400574, | |
| "eval_runtime": 55.3592, | |
| "eval_samples_per_second": 79.571, | |
| "eval_steps_per_second": 2.493, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 45.37205081669691, | |
| "grad_norm": 1.936480164527893, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.3996, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 45.37205081669691, | |
| "eval_accuracy": 0.8945571305505521, | |
| "eval_loss": 0.4623182713985443, | |
| "eval_runtime": 53.3019, | |
| "eval_samples_per_second": 82.642, | |
| "eval_steps_per_second": 2.589, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 46.27949183303085, | |
| "grad_norm": 1.907658576965332, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 0.3979, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 46.27949183303085, | |
| "eval_accuracy": 0.8960829529232748, | |
| "eval_loss": 0.4556325376033783, | |
| "eval_runtime": 54.1617, | |
| "eval_samples_per_second": 81.331, | |
| "eval_steps_per_second": 2.548, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 47.186932849364794, | |
| "grad_norm": 1.9181513786315918, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.391, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 47.186932849364794, | |
| "eval_accuracy": 0.8958657660824587, | |
| "eval_loss": 0.4613765776157379, | |
| "eval_runtime": 55.5872, | |
| "eval_samples_per_second": 79.245, | |
| "eval_steps_per_second": 2.483, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 48.09437386569873, | |
| "grad_norm": 1.6843451261520386, | |
| "learning_rate": 7.35e-05, | |
| "loss": 0.391, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 48.09437386569873, | |
| "eval_accuracy": 0.89564026034286, | |
| "eval_loss": 0.45949628949165344, | |
| "eval_runtime": 52.8113, | |
| "eval_samples_per_second": 83.41, | |
| "eval_steps_per_second": 2.613, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 49.00181488203267, | |
| "grad_norm": 1.7681550979614258, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.3874, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 49.00181488203267, | |
| "eval_accuracy": 0.8962522308149911, | |
| "eval_loss": 0.4545115828514099, | |
| "eval_runtime": 53.1381, | |
| "eval_samples_per_second": 82.897, | |
| "eval_steps_per_second": 2.597, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 49.90925589836661, | |
| "grad_norm": 1.8517777919769287, | |
| "learning_rate": 7.25e-05, | |
| "loss": 0.3835, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 49.90925589836661, | |
| "eval_accuracy": 0.896625333542615, | |
| "eval_loss": 0.45060068368911743, | |
| "eval_runtime": 52.9396, | |
| "eval_samples_per_second": 83.208, | |
| "eval_steps_per_second": 2.607, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 50.81669691470054, | |
| "grad_norm": 1.9447550773620605, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.3779, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 50.81669691470054, | |
| "eval_accuracy": 0.8974899929361903, | |
| "eval_loss": 0.4529257118701935, | |
| "eval_runtime": 56.9809, | |
| "eval_samples_per_second": 77.307, | |
| "eval_steps_per_second": 2.422, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 51.724137931034484, | |
| "grad_norm": 1.7611163854599, | |
| "learning_rate": 7.15e-05, | |
| "loss": 0.3783, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 51.724137931034484, | |
| "eval_accuracy": 0.8984791687632082, | |
| "eval_loss": 0.44696977734565735, | |
| "eval_runtime": 53.9391, | |
| "eval_samples_per_second": 81.666, | |
| "eval_steps_per_second": 2.558, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 52.63157894736842, | |
| "grad_norm": 1.749190092086792, | |
| "learning_rate": 7.1e-05, | |
| "loss": 0.3727, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 52.63157894736842, | |
| "eval_accuracy": 0.8978207320708537, | |
| "eval_loss": 0.4506888687610626, | |
| "eval_runtime": 53.5362, | |
| "eval_samples_per_second": 82.281, | |
| "eval_steps_per_second": 2.578, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 53.53901996370236, | |
| "grad_norm": 1.8609730005264282, | |
| "learning_rate": 7.05e-05, | |
| "loss": 0.3705, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 53.53901996370236, | |
| "eval_accuracy": 0.8976168077767325, | |
| "eval_loss": 0.4500649869441986, | |
| "eval_runtime": 53.4168, | |
| "eval_samples_per_second": 82.465, | |
| "eval_steps_per_second": 2.583, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 54.4464609800363, | |
| "grad_norm": 1.8506393432617188, | |
| "learning_rate": 7e-05, | |
| "loss": 0.3719, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 54.4464609800363, | |
| "eval_accuracy": 0.8993505575402249, | |
| "eval_loss": 0.44908007979393005, | |
| "eval_runtime": 57.767, | |
| "eval_samples_per_second": 76.255, | |
| "eval_steps_per_second": 2.389, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 55.35390199637023, | |
| "grad_norm": 1.8105406761169434, | |
| "learning_rate": 6.95e-05, | |
| "loss": 0.3684, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 55.35390199637023, | |
| "eval_accuracy": 0.8994710323502174, | |
| "eval_loss": 0.44289711117744446, | |
| "eval_runtime": 53.0941, | |
| "eval_samples_per_second": 82.966, | |
| "eval_steps_per_second": 2.599, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 56.261343012704174, | |
| "grad_norm": 1.9500548839569092, | |
| "learning_rate": 6.9e-05, | |
| "loss": 0.3621, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 56.261343012704174, | |
| "eval_accuracy": 0.8995646091699282, | |
| "eval_loss": 0.44326120615005493, | |
| "eval_runtime": 57.7269, | |
| "eval_samples_per_second": 76.308, | |
| "eval_steps_per_second": 2.391, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 57.168784029038115, | |
| "grad_norm": 1.771316647529602, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 0.3639, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 57.168784029038115, | |
| "eval_accuracy": 0.8996551918347424, | |
| "eval_loss": 0.4400934875011444, | |
| "eval_runtime": 59.1245, | |
| "eval_samples_per_second": 74.504, | |
| "eval_steps_per_second": 2.334, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 58.07622504537205, | |
| "grad_norm": 1.9457340240478516, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.3603, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 58.07622504537205, | |
| "eval_accuracy": 0.8985898806146979, | |
| "eval_loss": 0.4473365545272827, | |
| "eval_runtime": 53.3011, | |
| "eval_samples_per_second": 82.644, | |
| "eval_steps_per_second": 2.589, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 58.98366606170599, | |
| "grad_norm": 1.8146084547042847, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 0.3568, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 58.98366606170599, | |
| "eval_accuracy": 0.9005719064701833, | |
| "eval_loss": 0.4383050501346588, | |
| "eval_runtime": 57.7373, | |
| "eval_samples_per_second": 76.294, | |
| "eval_steps_per_second": 2.39, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 59.89110707803993, | |
| "grad_norm": 1.809646725654602, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.3516, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 59.89110707803993, | |
| "eval_accuracy": 0.9000993445757237, | |
| "eval_loss": 0.44284284114837646, | |
| "eval_runtime": 53.187, | |
| "eval_samples_per_second": 82.821, | |
| "eval_steps_per_second": 2.595, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 60.798548094373864, | |
| "grad_norm": 1.7659413814544678, | |
| "learning_rate": 6.65e-05, | |
| "loss": 0.3531, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 60.798548094373864, | |
| "eval_accuracy": 0.8994799933629496, | |
| "eval_loss": 0.44589298963546753, | |
| "eval_runtime": 53.0032, | |
| "eval_samples_per_second": 83.108, | |
| "eval_steps_per_second": 2.604, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 61.705989110707804, | |
| "grad_norm": 1.636080265045166, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.3499, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 61.705989110707804, | |
| "eval_accuracy": 0.8999748562089449, | |
| "eval_loss": 0.4367033839225769, | |
| "eval_runtime": 53.0566, | |
| "eval_samples_per_second": 83.025, | |
| "eval_steps_per_second": 2.601, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 62.613430127041745, | |
| "grad_norm": 1.8669129610061646, | |
| "learning_rate": 6.55e-05, | |
| "loss": 0.3489, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 62.613430127041745, | |
| "eval_accuracy": 0.9013975155279503, | |
| "eval_loss": 0.4371834695339203, | |
| "eval_runtime": 53.1608, | |
| "eval_samples_per_second": 82.862, | |
| "eval_steps_per_second": 2.596, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 63.52087114337568, | |
| "grad_norm": 1.9811877012252808, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.3429, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 63.52087114337568, | |
| "eval_accuracy": 0.9018210634557753, | |
| "eval_loss": 0.4384971857070923, | |
| "eval_runtime": 57.6259, | |
| "eval_samples_per_second": 76.441, | |
| "eval_steps_per_second": 2.395, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 64.42831215970962, | |
| "grad_norm": 1.7895385026931763, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 0.3415, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 64.42831215970962, | |
| "eval_accuracy": 0.9028058283836883, | |
| "eval_loss": 0.43259307742118835, | |
| "eval_runtime": 53.0105, | |
| "eval_samples_per_second": 83.097, | |
| "eval_steps_per_second": 2.603, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 65.33575317604355, | |
| "grad_norm": 1.9592262506484985, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.3402, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 65.33575317604355, | |
| "eval_accuracy": 0.901298823973929, | |
| "eval_loss": 0.4333614110946655, | |
| "eval_runtime": 52.9956, | |
| "eval_samples_per_second": 83.12, | |
| "eval_steps_per_second": 2.604, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 66.2431941923775, | |
| "grad_norm": 1.69992196559906, | |
| "learning_rate": 6.35e-05, | |
| "loss": 0.3349, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 66.2431941923775, | |
| "eval_accuracy": 0.90217415310253, | |
| "eval_loss": 0.430584579706192, | |
| "eval_runtime": 52.9983, | |
| "eval_samples_per_second": 83.116, | |
| "eval_steps_per_second": 2.604, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 67.15063520871144, | |
| "grad_norm": 1.70908522605896, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.3387, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 67.15063520871144, | |
| "eval_accuracy": 0.9022541035115501, | |
| "eval_loss": 0.4351217746734619, | |
| "eval_runtime": 53.4219, | |
| "eval_samples_per_second": 82.457, | |
| "eval_steps_per_second": 2.583, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 68.05807622504537, | |
| "grad_norm": 1.7078979015350342, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.3328, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 68.05807622504537, | |
| "eval_accuracy": 0.9027885079291, | |
| "eval_loss": 0.43064776062965393, | |
| "eval_runtime": 52.9742, | |
| "eval_samples_per_second": 83.154, | |
| "eval_steps_per_second": 2.605, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 68.96551724137932, | |
| "grad_norm": 1.6770516633987427, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.33, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 68.96551724137932, | |
| "eval_accuracy": 0.9039243367993435, | |
| "eval_loss": 0.42335009574890137, | |
| "eval_runtime": 52.9527, | |
| "eval_samples_per_second": 83.187, | |
| "eval_steps_per_second": 2.606, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 69.87295825771325, | |
| "grad_norm": 1.7739548683166504, | |
| "learning_rate": 6.15e-05, | |
| "loss": 0.3291, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 69.87295825771325, | |
| "eval_accuracy": 0.9030174309735959, | |
| "eval_loss": 0.42873483896255493, | |
| "eval_runtime": 52.9154, | |
| "eval_samples_per_second": 83.246, | |
| "eval_steps_per_second": 2.608, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 70.78039927404718, | |
| "grad_norm": 1.9532561302185059, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.3288, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 70.78039927404718, | |
| "eval_accuracy": 0.9036126447268178, | |
| "eval_loss": 0.42941179871559143, | |
| "eval_runtime": 53.0126, | |
| "eval_samples_per_second": 83.093, | |
| "eval_steps_per_second": 2.603, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 71.68784029038113, | |
| "grad_norm": 1.9629998207092285, | |
| "learning_rate": 6.05e-05, | |
| "loss": 0.3255, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 71.68784029038113, | |
| "eval_accuracy": 0.904484644880421, | |
| "eval_loss": 0.4233216643333435, | |
| "eval_runtime": 52.8794, | |
| "eval_samples_per_second": 83.303, | |
| "eval_steps_per_second": 2.61, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 72.59528130671507, | |
| "grad_norm": 1.9358775615692139, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3248, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 72.59528130671507, | |
| "eval_accuracy": 0.9041070159565976, | |
| "eval_loss": 0.4286067485809326, | |
| "eval_runtime": 52.8963, | |
| "eval_samples_per_second": 83.276, | |
| "eval_steps_per_second": 2.609, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 73.502722323049, | |
| "grad_norm": 1.6557801961898804, | |
| "learning_rate": 5.95e-05, | |
| "loss": 0.3203, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 73.502722323049, | |
| "eval_accuracy": 0.9040081056486239, | |
| "eval_loss": 0.4276265799999237, | |
| "eval_runtime": 52.8201, | |
| "eval_samples_per_second": 83.396, | |
| "eval_steps_per_second": 2.613, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 74.41016333938293, | |
| "grad_norm": 1.58525812625885, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.3218, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 74.41016333938293, | |
| "eval_accuracy": 0.9048278199949151, | |
| "eval_loss": 0.4280424118041992, | |
| "eval_runtime": 53.6976, | |
| "eval_samples_per_second": 82.034, | |
| "eval_steps_per_second": 2.57, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 75.31760435571688, | |
| "grad_norm": 1.5357856750488281, | |
| "learning_rate": 5.85e-05, | |
| "loss": 0.3198, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 75.31760435571688, | |
| "eval_accuracy": 0.904816966551977, | |
| "eval_loss": 0.41927599906921387, | |
| "eval_runtime": 52.6237, | |
| "eval_samples_per_second": 83.708, | |
| "eval_steps_per_second": 2.622, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 76.22504537205081, | |
| "grad_norm": 1.9634026288986206, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.3149, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 76.22504537205081, | |
| "eval_accuracy": 0.9031350344081517, | |
| "eval_loss": 0.4304438531398773, | |
| "eval_runtime": 52.8667, | |
| "eval_samples_per_second": 83.323, | |
| "eval_steps_per_second": 2.61, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 77.13248638838475, | |
| "grad_norm": 1.563607096672058, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.3144, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 77.13248638838475, | |
| "eval_accuracy": 0.9048774759257809, | |
| "eval_loss": 0.427058607339859, | |
| "eval_runtime": 52.712, | |
| "eval_samples_per_second": 83.567, | |
| "eval_steps_per_second": 2.618, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 78.0399274047187, | |
| "grad_norm": 1.871159553527832, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.3103, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 78.0399274047187, | |
| "eval_accuracy": 0.9048358855975869, | |
| "eval_loss": 0.4271075129508972, | |
| "eval_runtime": 52.7904, | |
| "eval_samples_per_second": 83.443, | |
| "eval_steps_per_second": 2.614, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 78.94736842105263, | |
| "grad_norm": 1.61452054977417, | |
| "learning_rate": 5.65e-05, | |
| "loss": 0.3102, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 78.94736842105263, | |
| "eval_accuracy": 0.9049740443605474, | |
| "eval_loss": 0.42336586117744446, | |
| "eval_runtime": 52.875, | |
| "eval_samples_per_second": 83.31, | |
| "eval_steps_per_second": 2.61, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 79.85480943738656, | |
| "grad_norm": 1.8801889419555664, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.307, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 79.85480943738656, | |
| "eval_accuracy": 0.9055931173260359, | |
| "eval_loss": 0.42033708095550537, | |
| "eval_runtime": 52.627, | |
| "eval_samples_per_second": 83.702, | |
| "eval_steps_per_second": 2.622, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 80.76225045372051, | |
| "grad_norm": 1.5314077138900757, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 0.3037, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 80.76225045372051, | |
| "eval_accuracy": 0.9061196499462085, | |
| "eval_loss": 0.4253558814525604, | |
| "eval_runtime": 52.7114, | |
| "eval_samples_per_second": 83.568, | |
| "eval_steps_per_second": 2.618, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 81.66969147005445, | |
| "grad_norm": 1.7618950605392456, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.3016, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 81.66969147005445, | |
| "eval_accuracy": 0.9062959555947709, | |
| "eval_loss": 0.4181654453277588, | |
| "eval_runtime": 52.7251, | |
| "eval_samples_per_second": 83.547, | |
| "eval_steps_per_second": 2.617, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 82.57713248638838, | |
| "grad_norm": 1.913796067237854, | |
| "learning_rate": 5.45e-05, | |
| "loss": 0.303, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 82.57713248638838, | |
| "eval_accuracy": 0.9072205041995722, | |
| "eval_loss": 0.41656142473220825, | |
| "eval_runtime": 52.7847, | |
| "eval_samples_per_second": 83.452, | |
| "eval_steps_per_second": 2.614, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 83.48457350272233, | |
| "grad_norm": 1.803902506828308, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.3028, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 83.48457350272233, | |
| "eval_accuracy": 0.9070220964556146, | |
| "eval_loss": 0.4230930507183075, | |
| "eval_runtime": 54.7506, | |
| "eval_samples_per_second": 80.456, | |
| "eval_steps_per_second": 2.521, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 84.39201451905626, | |
| "grad_norm": 1.7706644535064697, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 0.2986, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 84.39201451905626, | |
| "eval_accuracy": 0.9074832026084252, | |
| "eval_loss": 0.4140300452709198, | |
| "eval_runtime": 52.7024, | |
| "eval_samples_per_second": 83.583, | |
| "eval_steps_per_second": 2.618, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 85.2994555353902, | |
| "grad_norm": 1.9356876611709595, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 0.2966, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 85.2994555353902, | |
| "eval_accuracy": 0.9071740298423052, | |
| "eval_loss": 0.4191630482673645, | |
| "eval_runtime": 52.7256, | |
| "eval_samples_per_second": 83.546, | |
| "eval_steps_per_second": 2.617, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 86.20689655172414, | |
| "grad_norm": 1.8884636163711548, | |
| "learning_rate": 5.25e-05, | |
| "loss": 0.2959, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 86.20689655172414, | |
| "eval_accuracy": 0.9080046323395202, | |
| "eval_loss": 0.4184423089027405, | |
| "eval_runtime": 52.5803, | |
| "eval_samples_per_second": 83.777, | |
| "eval_steps_per_second": 2.625, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 87.11433756805808, | |
| "grad_norm": 1.7885215282440186, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.2943, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 87.11433756805808, | |
| "eval_accuracy": 0.9073709179447276, | |
| "eval_loss": 0.4168856143951416, | |
| "eval_runtime": 52.6737, | |
| "eval_samples_per_second": 83.628, | |
| "eval_steps_per_second": 2.62, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 88.02177858439201, | |
| "grad_norm": 1.6675046682357788, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 0.2932, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 88.02177858439201, | |
| "eval_accuracy": 0.9065622782059254, | |
| "eval_loss": 0.42474210262298584, | |
| "eval_runtime": 57.2893, | |
| "eval_samples_per_second": 76.89, | |
| "eval_steps_per_second": 2.409, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 88.92921960072596, | |
| "grad_norm": 1.7693278789520264, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 0.2913, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 88.92921960072596, | |
| "eval_accuracy": 0.908334698713182, | |
| "eval_loss": 0.41496196389198303, | |
| "eval_runtime": 52.681, | |
| "eval_samples_per_second": 83.616, | |
| "eval_steps_per_second": 2.62, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 89.83666061705989, | |
| "grad_norm": 1.80568528175354, | |
| "learning_rate": 5.05e-05, | |
| "loss": 0.29, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 89.83666061705989, | |
| "eval_accuracy": 0.9067204115993917, | |
| "eval_loss": 0.42078179121017456, | |
| "eval_runtime": 52.5861, | |
| "eval_samples_per_second": 83.767, | |
| "eval_steps_per_second": 2.624, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 90.74410163339383, | |
| "grad_norm": 1.7286852598190308, | |
| "learning_rate": 5e-05, | |
| "loss": 0.288, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 90.74410163339383, | |
| "eval_accuracy": 0.908493707496287, | |
| "eval_loss": 0.41103261709213257, | |
| "eval_runtime": 52.6627, | |
| "eval_samples_per_second": 83.646, | |
| "eval_steps_per_second": 2.62, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 91.65154264972777, | |
| "grad_norm": 1.6064249277114868, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.2846, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 91.65154264972777, | |
| "eval_accuracy": 0.9089611693118373, | |
| "eval_loss": 0.4114561975002289, | |
| "eval_runtime": 52.5213, | |
| "eval_samples_per_second": 83.871, | |
| "eval_steps_per_second": 2.628, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 92.5589836660617, | |
| "grad_norm": 1.6957948207855225, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.2825, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 92.5589836660617, | |
| "eval_accuracy": 0.9070289798162255, | |
| "eval_loss": 0.41827893257141113, | |
| "eval_runtime": 52.6164, | |
| "eval_samples_per_second": 83.719, | |
| "eval_steps_per_second": 2.623, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 93.46642468239564, | |
| "grad_norm": 1.6073497533798218, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.2834, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 93.46642468239564, | |
| "eval_accuracy": 0.9093731751882107, | |
| "eval_loss": 0.40947192907333374, | |
| "eval_runtime": 55.8553, | |
| "eval_samples_per_second": 78.865, | |
| "eval_steps_per_second": 2.471, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 94.37386569872959, | |
| "grad_norm": 1.8224419355392456, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.2803, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 94.37386569872959, | |
| "eval_accuracy": 0.9078407463018748, | |
| "eval_loss": 0.418082594871521, | |
| "eval_runtime": 57.2453, | |
| "eval_samples_per_second": 76.95, | |
| "eval_steps_per_second": 2.411, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 95.28130671506352, | |
| "grad_norm": 1.9654055833816528, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.2787, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 95.28130671506352, | |
| "eval_accuracy": 0.9089184717302816, | |
| "eval_loss": 0.4162246882915497, | |
| "eval_runtime": 57.2773, | |
| "eval_samples_per_second": 76.907, | |
| "eval_steps_per_second": 2.409, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 96.18874773139746, | |
| "grad_norm": 1.7956877946853638, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.278, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 96.18874773139746, | |
| "eval_accuracy": 0.9096441756342786, | |
| "eval_loss": 0.40937647223472595, | |
| "eval_runtime": 57.2739, | |
| "eval_samples_per_second": 76.911, | |
| "eval_steps_per_second": 2.409, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 97.0961887477314, | |
| "grad_norm": 1.6677452325820923, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.2759, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 97.0961887477314, | |
| "eval_accuracy": 0.9090880598745718, | |
| "eval_loss": 0.4136127233505249, | |
| "eval_runtime": 52.9843, | |
| "eval_samples_per_second": 83.138, | |
| "eval_steps_per_second": 2.605, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 98.00362976406534, | |
| "grad_norm": 1.7029211521148682, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.2746, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 98.00362976406534, | |
| "eval_accuracy": 0.9085475626951355, | |
| "eval_loss": 0.4151366651058197, | |
| "eval_runtime": 55.2968, | |
| "eval_samples_per_second": 79.661, | |
| "eval_steps_per_second": 2.496, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 98.91107078039927, | |
| "grad_norm": 1.4931912422180176, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.2734, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 98.91107078039927, | |
| "eval_accuracy": 0.9087434484443874, | |
| "eval_loss": 0.4170074760913849, | |
| "eval_runtime": 52.6515, | |
| "eval_samples_per_second": 83.663, | |
| "eval_steps_per_second": 2.621, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 99.81851179673322, | |
| "grad_norm": 1.8849012851715088, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.2719, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 99.81851179673322, | |
| "eval_accuracy": 0.9087569026104417, | |
| "eval_loss": 0.41325053572654724, | |
| "eval_runtime": 52.6731, | |
| "eval_samples_per_second": 83.629, | |
| "eval_steps_per_second": 2.62, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 100.72595281306715, | |
| "grad_norm": 2.079172372817993, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.271, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 100.72595281306715, | |
| "eval_accuracy": 0.9103295110887096, | |
| "eval_loss": 0.41259288787841797, | |
| "eval_runtime": 52.6285, | |
| "eval_samples_per_second": 83.7, | |
| "eval_steps_per_second": 2.622, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 101.63339382940109, | |
| "grad_norm": 1.8264790773391724, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.267, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 101.63339382940109, | |
| "eval_accuracy": 0.9093908939634668, | |
| "eval_loss": 0.41436412930488586, | |
| "eval_runtime": 53.0842, | |
| "eval_samples_per_second": 82.981, | |
| "eval_steps_per_second": 2.6, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 102.54083484573503, | |
| "grad_norm": 1.6904346942901611, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.2682, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 102.54083484573503, | |
| "eval_accuracy": 0.9092624512952864, | |
| "eval_loss": 0.413330078125, | |
| "eval_runtime": 57.4299, | |
| "eval_samples_per_second": 76.702, | |
| "eval_steps_per_second": 2.403, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 103.44827586206897, | |
| "grad_norm": 1.7619383335113525, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.2681, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 103.44827586206897, | |
| "eval_accuracy": 0.9103495162664795, | |
| "eval_loss": 0.40912753343582153, | |
| "eval_runtime": 57.2635, | |
| "eval_samples_per_second": 76.925, | |
| "eval_steps_per_second": 2.41, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 104.3557168784029, | |
| "grad_norm": 1.7024521827697754, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.2644, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 104.3557168784029, | |
| "eval_accuracy": 0.9091305589286443, | |
| "eval_loss": 0.4176701605319977, | |
| "eval_runtime": 52.5316, | |
| "eval_samples_per_second": 83.854, | |
| "eval_steps_per_second": 2.627, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 105.26315789473684, | |
| "grad_norm": 1.936393141746521, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2621, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 105.26315789473684, | |
| "eval_accuracy": 0.909638402972615, | |
| "eval_loss": 0.4138263165950775, | |
| "eval_runtime": 57.3248, | |
| "eval_samples_per_second": 76.843, | |
| "eval_steps_per_second": 2.407, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 106.17059891107078, | |
| "grad_norm": 1.9783495664596558, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.2618, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 106.17059891107078, | |
| "eval_accuracy": 0.9093778717725997, | |
| "eval_loss": 0.41479504108428955, | |
| "eval_runtime": 52.6769, | |
| "eval_samples_per_second": 83.623, | |
| "eval_steps_per_second": 2.62, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 107.07803992740472, | |
| "grad_norm": 1.8036541938781738, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.2593, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 107.07803992740472, | |
| "eval_accuracy": 0.9108752466798229, | |
| "eval_loss": 0.407368004322052, | |
| "eval_runtime": 52.7623, | |
| "eval_samples_per_second": 83.488, | |
| "eval_steps_per_second": 2.616, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 107.98548094373865, | |
| "grad_norm": 1.7356771230697632, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.2586, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 107.98548094373865, | |
| "eval_accuracy": 0.9106982338220416, | |
| "eval_loss": 0.4060940444469452, | |
| "eval_runtime": 57.2822, | |
| "eval_samples_per_second": 76.9, | |
| "eval_steps_per_second": 2.409, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 108.8929219600726, | |
| "grad_norm": 1.9993195533752441, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2591, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 108.8929219600726, | |
| "eval_accuracy": 0.910394873244746, | |
| "eval_loss": 0.4131792187690735, | |
| "eval_runtime": 53.0, | |
| "eval_samples_per_second": 83.113, | |
| "eval_steps_per_second": 2.604, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 109.80036297640653, | |
| "grad_norm": 1.8162901401519775, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.2558, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 109.80036297640653, | |
| "eval_accuracy": 0.910258865637902, | |
| "eval_loss": 0.4147132933139801, | |
| "eval_runtime": 55.8856, | |
| "eval_samples_per_second": 78.822, | |
| "eval_steps_per_second": 2.469, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 110.70780399274047, | |
| "grad_norm": 1.8106731176376343, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.2541, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 110.70780399274047, | |
| "eval_accuracy": 0.9114573371669734, | |
| "eval_loss": 0.40593624114990234, | |
| "eval_runtime": 52.8519, | |
| "eval_samples_per_second": 83.346, | |
| "eval_steps_per_second": 2.611, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 111.61524500907441, | |
| "grad_norm": 1.748769998550415, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.2556, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 111.61524500907441, | |
| "eval_accuracy": 0.9115416885324719, | |
| "eval_loss": 0.4094337522983551, | |
| "eval_runtime": 52.8285, | |
| "eval_samples_per_second": 83.383, | |
| "eval_steps_per_second": 2.612, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 112.52268602540835, | |
| "grad_norm": 1.6545246839523315, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.25, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 112.52268602540835, | |
| "eval_accuracy": 0.9117101451094991, | |
| "eval_loss": 0.4004589915275574, | |
| "eval_runtime": 52.6819, | |
| "eval_samples_per_second": 83.615, | |
| "eval_steps_per_second": 2.619, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 113.43012704174228, | |
| "grad_norm": 1.9466238021850586, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2492, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 113.43012704174228, | |
| "eval_accuracy": 0.9121271025446342, | |
| "eval_loss": 0.4025409519672394, | |
| "eval_runtime": 52.7085, | |
| "eval_samples_per_second": 83.573, | |
| "eval_steps_per_second": 2.618, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 114.33756805807623, | |
| "grad_norm": 1.8718467950820923, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.2505, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 114.33756805807623, | |
| "eval_accuracy": 0.9116951165625715, | |
| "eval_loss": 0.40717950463294983, | |
| "eval_runtime": 52.6815, | |
| "eval_samples_per_second": 83.616, | |
| "eval_steps_per_second": 2.62, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 115.24500907441016, | |
| "grad_norm": 1.6669822931289673, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.2477, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 115.24500907441016, | |
| "eval_accuracy": 0.9123295778283549, | |
| "eval_loss": 0.40226927399635315, | |
| "eval_runtime": 53.2555, | |
| "eval_samples_per_second": 82.714, | |
| "eval_steps_per_second": 2.591, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 116.1524500907441, | |
| "grad_norm": 1.631198525428772, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.2462, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 116.1524500907441, | |
| "eval_accuracy": 0.9121495562330304, | |
| "eval_loss": 0.4079442322254181, | |
| "eval_runtime": 52.7267, | |
| "eval_samples_per_second": 83.544, | |
| "eval_steps_per_second": 2.617, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 117.05989110707804, | |
| "grad_norm": 1.7638319730758667, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.2472, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 117.05989110707804, | |
| "eval_accuracy": 0.9112412273671573, | |
| "eval_loss": 0.40657439827919006, | |
| "eval_runtime": 52.7752, | |
| "eval_samples_per_second": 83.467, | |
| "eval_steps_per_second": 2.615, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 117.96733212341198, | |
| "grad_norm": 1.821175217628479, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2436, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 117.96733212341198, | |
| "eval_accuracy": 0.9124436813445042, | |
| "eval_loss": 0.40668636560440063, | |
| "eval_runtime": 52.7669, | |
| "eval_samples_per_second": 83.48, | |
| "eval_steps_per_second": 2.615, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 118.87477313974591, | |
| "grad_norm": 1.8072514533996582, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.2432, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 118.87477313974591, | |
| "eval_accuracy": 0.9114819864290975, | |
| "eval_loss": 0.4095401167869568, | |
| "eval_runtime": 52.6486, | |
| "eval_samples_per_second": 83.668, | |
| "eval_steps_per_second": 2.621, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 119.78221415607986, | |
| "grad_norm": 1.8061636686325073, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.2406, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 119.78221415607986, | |
| "eval_accuracy": 0.9118786110814227, | |
| "eval_loss": 0.4098331332206726, | |
| "eval_runtime": 52.7138, | |
| "eval_samples_per_second": 83.564, | |
| "eval_steps_per_second": 2.618, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 120.6896551724138, | |
| "grad_norm": 1.9020588397979736, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.2421, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 120.6896551724138, | |
| "eval_accuracy": 0.9126196512325608, | |
| "eval_loss": 0.4041764736175537, | |
| "eval_runtime": 52.7206, | |
| "eval_samples_per_second": 83.554, | |
| "eval_steps_per_second": 2.618, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 121.59709618874773, | |
| "grad_norm": 1.8141471147537231, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.2407, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 121.59709618874773, | |
| "eval_accuracy": 0.9133736295696817, | |
| "eval_loss": 0.40654563903808594, | |
| "eval_runtime": 52.5772, | |
| "eval_samples_per_second": 83.782, | |
| "eval_steps_per_second": 2.625, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 122.50453720508168, | |
| "grad_norm": 1.9479206800460815, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2392, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 122.50453720508168, | |
| "eval_accuracy": 0.9131637376284198, | |
| "eval_loss": 0.4058144688606262, | |
| "eval_runtime": 57.3018, | |
| "eval_samples_per_second": 76.874, | |
| "eval_steps_per_second": 2.408, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 123.41197822141561, | |
| "grad_norm": 2.147027015686035, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.2377, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 123.41197822141561, | |
| "eval_accuracy": 0.9131377603207724, | |
| "eval_loss": 0.4089277386665344, | |
| "eval_runtime": 52.5836, | |
| "eval_samples_per_second": 83.771, | |
| "eval_steps_per_second": 2.624, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 124.31941923774954, | |
| "grad_norm": 1.95304536819458, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.2372, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 124.31941923774954, | |
| "eval_accuracy": 0.9145953679805113, | |
| "eval_loss": 0.4008789658546448, | |
| "eval_runtime": 52.5737, | |
| "eval_samples_per_second": 83.787, | |
| "eval_steps_per_second": 2.625, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 125.22686025408349, | |
| "grad_norm": 1.8719152212142944, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.234, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 125.22686025408349, | |
| "eval_accuracy": 0.9142573977063571, | |
| "eval_loss": 0.4005224108695984, | |
| "eval_runtime": 54.591, | |
| "eval_samples_per_second": 80.691, | |
| "eval_steps_per_second": 2.528, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 126.13430127041742, | |
| "grad_norm": 1.9268224239349365, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.2342, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 126.13430127041742, | |
| "eval_accuracy": 0.9132593660123727, | |
| "eval_loss": 0.41198381781578064, | |
| "eval_runtime": 52.5548, | |
| "eval_samples_per_second": 83.817, | |
| "eval_steps_per_second": 2.626, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 127.04174228675136, | |
| "grad_norm": 1.9191150665283203, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2348, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 127.04174228675136, | |
| "eval_accuracy": 0.913298348179974, | |
| "eval_loss": 0.40492549538612366, | |
| "eval_runtime": 52.6333, | |
| "eval_samples_per_second": 83.692, | |
| "eval_steps_per_second": 2.622, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 127.9491833030853, | |
| "grad_norm": 1.5569913387298584, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.2321, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 127.9491833030853, | |
| "eval_accuracy": 0.9125856672390897, | |
| "eval_loss": 0.4051525592803955, | |
| "eval_runtime": 52.6574, | |
| "eval_samples_per_second": 83.654, | |
| "eval_steps_per_second": 2.621, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 128.85662431941924, | |
| "grad_norm": 1.7746883630752563, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.2294, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 128.85662431941924, | |
| "eval_accuracy": 0.9139568516478013, | |
| "eval_loss": 0.4073280692100525, | |
| "eval_runtime": 52.6696, | |
| "eval_samples_per_second": 83.635, | |
| "eval_steps_per_second": 2.62, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 129.76406533575317, | |
| "grad_norm": 1.4891724586486816, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.2285, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 129.76406533575317, | |
| "eval_accuracy": 0.9131215447858751, | |
| "eval_loss": 0.40580666065216064, | |
| "eval_runtime": 52.7507, | |
| "eval_samples_per_second": 83.506, | |
| "eval_steps_per_second": 2.616, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 130.6715063520871, | |
| "grad_norm": 1.5842249393463135, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.2285, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 130.6715063520871, | |
| "eval_accuracy": 0.9142407479255579, | |
| "eval_loss": 0.4018247723579407, | |
| "eval_runtime": 54.7129, | |
| "eval_samples_per_second": 80.511, | |
| "eval_steps_per_second": 2.522, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 131.57894736842104, | |
| "grad_norm": 1.8539658784866333, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2285, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 131.57894736842104, | |
| "eval_accuracy": 0.914012928454821, | |
| "eval_loss": 0.4047853946685791, | |
| "eval_runtime": 53.2071, | |
| "eval_samples_per_second": 82.79, | |
| "eval_steps_per_second": 2.594, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 132.486388384755, | |
| "grad_norm": 1.4111963510513306, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.227, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 132.486388384755, | |
| "eval_accuracy": 0.913627272698819, | |
| "eval_loss": 0.4063122570514679, | |
| "eval_runtime": 54.764, | |
| "eval_samples_per_second": 80.436, | |
| "eval_steps_per_second": 2.52, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 133.39382940108894, | |
| "grad_norm": 1.7000839710235596, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.2227, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 133.39382940108894, | |
| "eval_accuracy": 0.9136585735388811, | |
| "eval_loss": 0.40767940878868103, | |
| "eval_runtime": 52.6716, | |
| "eval_samples_per_second": 83.631, | |
| "eval_steps_per_second": 2.62, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 134.30127041742287, | |
| "grad_norm": 1.7322769165039062, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.2227, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 134.30127041742287, | |
| "eval_accuracy": 0.9146321198686937, | |
| "eval_loss": 0.40449175238609314, | |
| "eval_runtime": 57.3022, | |
| "eval_samples_per_second": 76.873, | |
| "eval_steps_per_second": 2.408, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 135.2087114337568, | |
| "grad_norm": 1.8843836784362793, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.2227, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 135.2087114337568, | |
| "eval_accuracy": 0.9153162670489917, | |
| "eval_loss": 0.3996308743953705, | |
| "eval_runtime": 52.9537, | |
| "eval_samples_per_second": 83.186, | |
| "eval_steps_per_second": 2.606, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 136.11615245009074, | |
| "grad_norm": 1.688589334487915, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2228, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 136.11615245009074, | |
| "eval_accuracy": 0.9148934837092731, | |
| "eval_loss": 0.4007312059402466, | |
| "eval_runtime": 52.7857, | |
| "eval_samples_per_second": 83.451, | |
| "eval_steps_per_second": 2.614, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 137.02359346642467, | |
| "grad_norm": 1.886564016342163, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.2222, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 137.02359346642467, | |
| "eval_accuracy": 0.9137650871178321, | |
| "eval_loss": 0.40756621956825256, | |
| "eval_runtime": 52.9485, | |
| "eval_samples_per_second": 83.194, | |
| "eval_steps_per_second": 2.606, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 137.93103448275863, | |
| "grad_norm": 1.569810152053833, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.2186, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 137.93103448275863, | |
| "eval_accuracy": 0.9148974119075408, | |
| "eval_loss": 0.4073057770729065, | |
| "eval_runtime": 52.8219, | |
| "eval_samples_per_second": 83.393, | |
| "eval_steps_per_second": 2.613, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 138.83847549909257, | |
| "grad_norm": 1.5704463720321655, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.2189, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 138.83847549909257, | |
| "eval_accuracy": 0.9138604511070603, | |
| "eval_loss": 0.40438133478164673, | |
| "eval_runtime": 52.8176, | |
| "eval_samples_per_second": 83.4, | |
| "eval_steps_per_second": 2.613, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 139.7459165154265, | |
| "grad_norm": 1.7239934206008911, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.2171, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 139.7459165154265, | |
| "eval_accuracy": 0.9152970248903189, | |
| "eval_loss": 0.4021734297275543, | |
| "eval_runtime": 53.338, | |
| "eval_samples_per_second": 82.587, | |
| "eval_steps_per_second": 2.587, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 140.65335753176043, | |
| "grad_norm": 1.79320228099823, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2167, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 140.65335753176043, | |
| "eval_accuracy": 0.9152779126251601, | |
| "eval_loss": 0.3990631401538849, | |
| "eval_runtime": 53.0839, | |
| "eval_samples_per_second": 82.982, | |
| "eval_steps_per_second": 2.6, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 141.56079854809437, | |
| "grad_norm": 1.6768089532852173, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.2164, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 141.56079854809437, | |
| "eval_accuracy": 0.9149024426267115, | |
| "eval_loss": 0.4050694704055786, | |
| "eval_runtime": 53.0695, | |
| "eval_samples_per_second": 83.004, | |
| "eval_steps_per_second": 2.6, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 142.4682395644283, | |
| "grad_norm": 1.9307670593261719, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.2154, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 142.4682395644283, | |
| "eval_accuracy": 0.9162203399408525, | |
| "eval_loss": 0.3998095691204071, | |
| "eval_runtime": 53.1001, | |
| "eval_samples_per_second": 82.956, | |
| "eval_steps_per_second": 2.599, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 143.37568058076226, | |
| "grad_norm": 1.617890477180481, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.215, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 143.37568058076226, | |
| "eval_accuracy": 0.9160746394397012, | |
| "eval_loss": 0.40022924542427063, | |
| "eval_runtime": 53.1216, | |
| "eval_samples_per_second": 82.923, | |
| "eval_steps_per_second": 2.598, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 144.2831215970962, | |
| "grad_norm": 2.0667710304260254, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.2126, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 144.2831215970962, | |
| "eval_accuracy": 0.9151828126910668, | |
| "eval_loss": 0.4023064076900482, | |
| "eval_runtime": 53.0659, | |
| "eval_samples_per_second": 83.01, | |
| "eval_steps_per_second": 2.601, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 145.19056261343013, | |
| "grad_norm": 1.772654414176941, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2134, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 145.19056261343013, | |
| "eval_accuracy": 0.9151423035495024, | |
| "eval_loss": 0.4016391932964325, | |
| "eval_runtime": 52.86, | |
| "eval_samples_per_second": 83.333, | |
| "eval_steps_per_second": 2.611, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 146.09800362976407, | |
| "grad_norm": 1.6949107646942139, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.2121, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 146.09800362976407, | |
| "eval_accuracy": 0.9157573098498856, | |
| "eval_loss": 0.40052202343940735, | |
| "eval_runtime": 52.9508, | |
| "eval_samples_per_second": 83.19, | |
| "eval_steps_per_second": 2.606, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 147.005444646098, | |
| "grad_norm": 1.7470875978469849, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.2118, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 147.005444646098, | |
| "eval_accuracy": 0.9147493650748915, | |
| "eval_loss": 0.4061746895313263, | |
| "eval_runtime": 53.021, | |
| "eval_samples_per_second": 83.08, | |
| "eval_steps_per_second": 2.603, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 147.91288566243193, | |
| "grad_norm": 1.6520947217941284, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.2092, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 147.91288566243193, | |
| "eval_accuracy": 0.9153819887159277, | |
| "eval_loss": 0.4039769172668457, | |
| "eval_runtime": 52.9737, | |
| "eval_samples_per_second": 83.155, | |
| "eval_steps_per_second": 2.605, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 148.82032667876587, | |
| "grad_norm": 1.625849962234497, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.2071, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 148.82032667876587, | |
| "eval_accuracy": 0.9155334497970264, | |
| "eval_loss": 0.4043760895729065, | |
| "eval_runtime": 52.9528, | |
| "eval_samples_per_second": 83.187, | |
| "eval_steps_per_second": 2.606, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 149.72776769509983, | |
| "grad_norm": 1.706663727760315, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2049, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 149.72776769509983, | |
| "eval_accuracy": 0.9146216897066012, | |
| "eval_loss": 0.4076862335205078, | |
| "eval_runtime": 53.5645, | |
| "eval_samples_per_second": 82.237, | |
| "eval_steps_per_second": 2.576, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 150.63520871143376, | |
| "grad_norm": 1.9129126071929932, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.2072, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 150.63520871143376, | |
| "eval_accuracy": 0.9153711232101976, | |
| "eval_loss": 0.39911210536956787, | |
| "eval_runtime": 53.0193, | |
| "eval_samples_per_second": 83.083, | |
| "eval_steps_per_second": 2.603, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 151.5426497277677, | |
| "grad_norm": 1.7741316556930542, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.2048, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 151.5426497277677, | |
| "eval_accuracy": 0.9165385170632505, | |
| "eval_loss": 0.3991451859474182, | |
| "eval_runtime": 53.0232, | |
| "eval_samples_per_second": 83.077, | |
| "eval_steps_per_second": 2.603, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 152.45009074410163, | |
| "grad_norm": 1.6072008609771729, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2064, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 152.45009074410163, | |
| "eval_accuracy": 0.9149193775458487, | |
| "eval_loss": 0.40828338265419006, | |
| "eval_runtime": 52.9412, | |
| "eval_samples_per_second": 83.206, | |
| "eval_steps_per_second": 2.607, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 153.35753176043556, | |
| "grad_norm": 1.7185778617858887, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.2061, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 153.35753176043556, | |
| "eval_accuracy": 0.9158725538979593, | |
| "eval_loss": 0.40011066198349, | |
| "eval_runtime": 52.978, | |
| "eval_samples_per_second": 83.148, | |
| "eval_steps_per_second": 2.605, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 154.2649727767695, | |
| "grad_norm": 1.7340868711471558, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.202, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 154.2649727767695, | |
| "eval_accuracy": 0.9167605678134148, | |
| "eval_loss": 0.3951858580112457, | |
| "eval_runtime": 53.0994, | |
| "eval_samples_per_second": 82.958, | |
| "eval_steps_per_second": 2.599, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 155.17241379310346, | |
| "grad_norm": 1.725895881652832, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.2007, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 155.17241379310346, | |
| "eval_accuracy": 0.9168939812952535, | |
| "eval_loss": 0.3987417221069336, | |
| "eval_runtime": 54.957, | |
| "eval_samples_per_second": 80.154, | |
| "eval_steps_per_second": 2.511, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 156.0798548094374, | |
| "grad_norm": 1.5828238725662231, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.2015, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 156.0798548094374, | |
| "eval_accuracy": 0.917391414109539, | |
| "eval_loss": 0.4005061388015747, | |
| "eval_runtime": 52.8456, | |
| "eval_samples_per_second": 83.356, | |
| "eval_steps_per_second": 2.611, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 156.98729582577133, | |
| "grad_norm": 1.631608247756958, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.2017, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 156.98729582577133, | |
| "eval_accuracy": 0.9161869769340681, | |
| "eval_loss": 0.40137505531311035, | |
| "eval_runtime": 53.0267, | |
| "eval_samples_per_second": 83.071, | |
| "eval_steps_per_second": 2.602, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 157.89473684210526, | |
| "grad_norm": 1.9344474077224731, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.1977, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 157.89473684210526, | |
| "eval_accuracy": 0.9169892177992084, | |
| "eval_loss": 0.39951831102371216, | |
| "eval_runtime": 52.9394, | |
| "eval_samples_per_second": 83.208, | |
| "eval_steps_per_second": 2.607, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 158.8021778584392, | |
| "grad_norm": 1.6796910762786865, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1994, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 158.8021778584392, | |
| "eval_accuracy": 0.9157372947418714, | |
| "eval_loss": 0.4002035856246948, | |
| "eval_runtime": 52.8869, | |
| "eval_samples_per_second": 83.291, | |
| "eval_steps_per_second": 2.609, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 159.70961887477313, | |
| "grad_norm": 1.5331308841705322, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1987, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 159.70961887477313, | |
| "eval_accuracy": 0.9159691578071264, | |
| "eval_loss": 0.40263810753822327, | |
| "eval_runtime": 57.4412, | |
| "eval_samples_per_second": 76.687, | |
| "eval_steps_per_second": 2.402, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 160.6170598911071, | |
| "grad_norm": 1.8451423645019531, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.1985, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 160.6170598911071, | |
| "eval_accuracy": 0.91600935818353, | |
| "eval_loss": 0.4028289318084717, | |
| "eval_runtime": 53.2779, | |
| "eval_samples_per_second": 82.68, | |
| "eval_steps_per_second": 2.59, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 161.52450090744102, | |
| "grad_norm": 1.6251318454742432, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.1976, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 161.52450090744102, | |
| "eval_accuracy": 0.9168077236536895, | |
| "eval_loss": 0.39661651849746704, | |
| "eval_runtime": 54.3943, | |
| "eval_samples_per_second": 80.983, | |
| "eval_steps_per_second": 2.537, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 162.43194192377496, | |
| "grad_norm": 1.906327486038208, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.1975, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 162.43194192377496, | |
| "eval_accuracy": 0.9169821754553538, | |
| "eval_loss": 0.3963495194911957, | |
| "eval_runtime": 53.0048, | |
| "eval_samples_per_second": 83.106, | |
| "eval_steps_per_second": 2.604, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 163.3393829401089, | |
| "grad_norm": 1.9544309377670288, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1963, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 163.3393829401089, | |
| "eval_accuracy": 0.9164625099202927, | |
| "eval_loss": 0.404565691947937, | |
| "eval_runtime": 53.9246, | |
| "eval_samples_per_second": 81.688, | |
| "eval_steps_per_second": 2.559, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 164.24682395644282, | |
| "grad_norm": 1.852169156074524, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.1963, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 164.24682395644282, | |
| "eval_accuracy": 0.9155449059728268, | |
| "eval_loss": 0.406656414270401, | |
| "eval_runtime": 52.9408, | |
| "eval_samples_per_second": 83.206, | |
| "eval_steps_per_second": 2.607, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 165.15426497277676, | |
| "grad_norm": 1.7243677377700806, | |
| "learning_rate": 9e-06, | |
| "loss": 0.1985, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 165.15426497277676, | |
| "eval_accuracy": 0.9178093382768805, | |
| "eval_loss": 0.39525070786476135, | |
| "eval_runtime": 52.9935, | |
| "eval_samples_per_second": 83.123, | |
| "eval_steps_per_second": 2.604, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 166.06170598911072, | |
| "grad_norm": 1.7043794393539429, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.1936, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 166.06170598911072, | |
| "eval_accuracy": 0.9167059652035036, | |
| "eval_loss": 0.3972921669483185, | |
| "eval_runtime": 52.9476, | |
| "eval_samples_per_second": 83.195, | |
| "eval_steps_per_second": 2.606, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 166.96914700544465, | |
| "grad_norm": 1.6746214628219604, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1937, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 166.96914700544465, | |
| "eval_accuracy": 0.9164150593646615, | |
| "eval_loss": 0.40151235461235046, | |
| "eval_runtime": 52.9555, | |
| "eval_samples_per_second": 83.183, | |
| "eval_steps_per_second": 2.606, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 167.8765880217786, | |
| "grad_norm": 1.724612832069397, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1936, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 167.8765880217786, | |
| "eval_accuracy": 0.9173433401523613, | |
| "eval_loss": 0.40141040086746216, | |
| "eval_runtime": 53.1703, | |
| "eval_samples_per_second": 82.847, | |
| "eval_steps_per_second": 2.595, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 168.78402903811252, | |
| "grad_norm": 1.8417091369628906, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.1923, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 168.78402903811252, | |
| "eval_accuracy": 0.9186817674871586, | |
| "eval_loss": 0.39400893449783325, | |
| "eval_runtime": 56.2134, | |
| "eval_samples_per_second": 78.362, | |
| "eval_steps_per_second": 2.455, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 169.69147005444646, | |
| "grad_norm": 1.7512474060058594, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.1933, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 169.69147005444646, | |
| "eval_accuracy": 0.9175796477495107, | |
| "eval_loss": 0.3983522951602936, | |
| "eval_runtime": 52.9057, | |
| "eval_samples_per_second": 83.261, | |
| "eval_steps_per_second": 2.608, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 170.5989110707804, | |
| "grad_norm": 1.4073301553726196, | |
| "learning_rate": 6e-06, | |
| "loss": 0.1935, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 170.5989110707804, | |
| "eval_accuracy": 0.9187447843680622, | |
| "eval_loss": 0.3936294913291931, | |
| "eval_runtime": 52.7794, | |
| "eval_samples_per_second": 83.461, | |
| "eval_steps_per_second": 2.615, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 171.50635208711435, | |
| "grad_norm": 1.71249520778656, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.1932, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 171.50635208711435, | |
| "eval_accuracy": 0.9183512090422428, | |
| "eval_loss": 0.3950323760509491, | |
| "eval_runtime": 52.9329, | |
| "eval_samples_per_second": 83.219, | |
| "eval_steps_per_second": 2.607, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 172.41379310344828, | |
| "grad_norm": 1.4874622821807861, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1915, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 172.41379310344828, | |
| "eval_accuracy": 0.9178856529139476, | |
| "eval_loss": 0.39599481225013733, | |
| "eval_runtime": 51.846, | |
| "eval_samples_per_second": 84.963, | |
| "eval_steps_per_second": 2.662, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 173.32123411978222, | |
| "grad_norm": 1.898534893989563, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.1891, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 173.32123411978222, | |
| "eval_accuracy": 0.918700790482899, | |
| "eval_loss": 0.38806891441345215, | |
| "eval_runtime": 53.2348, | |
| "eval_samples_per_second": 82.747, | |
| "eval_steps_per_second": 2.592, | |
| "step": 95500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 182, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.06414374804652e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |