| { | |
| "best_metric": 0.41527342389411354, | |
| "best_model_checkpoint": "/groups/claytonm/enoriega/kw_pubmed/kw_pubmed_5000_0.0003/checkpoint-375", | |
| "epoch": 3.407671998757571, | |
| "global_step": 525, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0002984415584415584, | |
| "loss": 5.7783, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0002964935064935065, | |
| "loss": 3.8225, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0002945454545454545, | |
| "loss": 3.3158, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.38301114594218044, | |
| "eval_loss": 3.8663249015808105, | |
| "eval_runtime": 16.5167, | |
| "eval_samples_per_second": 605.449, | |
| "eval_steps_per_second": 18.951, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0002925974025974026, | |
| "loss": 3.0324, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0002906493506493506, | |
| "loss": 2.8348, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0002887012987012987, | |
| "loss": 2.7061, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 0.38962904911180773, | |
| "eval_loss": 3.9842898845672607, | |
| "eval_runtime": 16.7828, | |
| "eval_samples_per_second": 595.847, | |
| "eval_steps_per_second": 18.65, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00028675324675324674, | |
| "loss": 2.6148, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00028480519480519477, | |
| "loss": 2.5071, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002828571428571428, | |
| "loss": 2.4347, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.40025252525252525, | |
| "eval_loss": 3.96134614944458, | |
| "eval_runtime": 16.7186, | |
| "eval_samples_per_second": 598.137, | |
| "eval_steps_per_second": 18.722, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0002809090909090909, | |
| "loss": 2.386, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002789610389610389, | |
| "loss": 2.3579, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.000277012987012987, | |
| "loss": 2.2818, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_accuracy": 0.4000348310693138, | |
| "eval_loss": 4.00076961517334, | |
| "eval_runtime": 16.4241, | |
| "eval_samples_per_second": 608.861, | |
| "eval_steps_per_second": 19.057, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002754545454545454, | |
| "loss": 2.2853, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002738961038961039, | |
| "loss": 2.4897, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002719480519480519, | |
| "loss": 2.2269, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.39672587948450017, | |
| "eval_loss": 4.018723964691162, | |
| "eval_runtime": 16.4842, | |
| "eval_samples_per_second": 606.64, | |
| "eval_steps_per_second": 18.988, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00027, | |
| "loss": 2.1896, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00026805194805194803, | |
| "loss": 2.2399, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00026610389610389607, | |
| "loss": 2.1639, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.39946882619296414, | |
| "eval_loss": 4.048261642456055, | |
| "eval_runtime": 16.4769, | |
| "eval_samples_per_second": 606.909, | |
| "eval_steps_per_second": 18.996, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0002641558441558441, | |
| "loss": 2.2196, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0002622077922077922, | |
| "loss": 2.1771, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0002602597402597402, | |
| "loss": 2.096, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.40103622431208635, | |
| "eval_loss": 4.030768871307373, | |
| "eval_runtime": 16.4452, | |
| "eval_samples_per_second": 608.079, | |
| "eval_steps_per_second": 19.033, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0002583116883116883, | |
| "loss": 2.0654, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00025636363636363633, | |
| "loss": 2.1061, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002544155844155844, | |
| "loss": 2.0405, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.3975095785440613, | |
| "eval_loss": 4.094750881195068, | |
| "eval_runtime": 16.44, | |
| "eval_samples_per_second": 608.272, | |
| "eval_steps_per_second": 19.039, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00025246753246753245, | |
| "loss": 2.0486, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0002505194805194805, | |
| "loss": 2.0271, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00024857142857142857, | |
| "loss": 2.0288, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.4073058167885754, | |
| "eval_loss": 4.073542594909668, | |
| "eval_runtime": 16.4284, | |
| "eval_samples_per_second": 608.702, | |
| "eval_steps_per_second": 19.052, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0002466233766233766, | |
| "loss": 1.9626, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.0002446753246753247, | |
| "loss": 1.9723, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0002427272727272727, | |
| "loss": 1.9442, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.4082636711947057, | |
| "eval_loss": 4.0442705154418945, | |
| "eval_runtime": 16.4893, | |
| "eval_samples_per_second": 606.453, | |
| "eval_steps_per_second": 18.982, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00024077922077922077, | |
| "loss": 2.0944, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0002388311688311688, | |
| "loss": 1.8005, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00023688311688311686, | |
| "loss": 1.7948, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.4029519331243469, | |
| "eval_loss": 4.067967414855957, | |
| "eval_runtime": 16.8157, | |
| "eval_samples_per_second": 594.681, | |
| "eval_steps_per_second": 18.614, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.00023493506493506492, | |
| "loss": 1.7747, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.00023298701298701298, | |
| "loss": 1.7695, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.000231038961038961, | |
| "loss": 1.7509, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_accuracy": 0.40434517589690006, | |
| "eval_loss": 4.150224685668945, | |
| "eval_runtime": 16.8831, | |
| "eval_samples_per_second": 592.308, | |
| "eval_steps_per_second": 18.539, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00022909090909090907, | |
| "loss": 1.708, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00022714285714285713, | |
| "loss": 1.699, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00022519480519480519, | |
| "loss": 1.706, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_accuracy": 0.40917798676419365, | |
| "eval_loss": 4.179904460906982, | |
| "eval_runtime": 16.415, | |
| "eval_samples_per_second": 609.2, | |
| "eval_steps_per_second": 19.068, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.00022324675324675322, | |
| "loss": 1.6846, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00022129870129870128, | |
| "loss": 1.6696, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00021935064935064933, | |
| "loss": 1.6689, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_accuracy": 0.4087861372344131, | |
| "eval_loss": 4.159827709197998, | |
| "eval_runtime": 16.4109, | |
| "eval_samples_per_second": 609.351, | |
| "eval_steps_per_second": 19.073, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.0002174025974025974, | |
| "loss": 1.687, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00021545454545454542, | |
| "loss": 1.65, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00021350649350649348, | |
| "loss": 1.6472, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_accuracy": 0.40599965168930685, | |
| "eval_loss": 4.144142150878906, | |
| "eval_runtime": 16.4801, | |
| "eval_samples_per_second": 606.792, | |
| "eval_steps_per_second": 18.993, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00021155844155844154, | |
| "loss": 1.634, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.0002096103896103896, | |
| "loss": 1.6136, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00020766233766233763, | |
| "loss": 1.6103, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.41266109369557646, | |
| "eval_loss": 4.268022537231445, | |
| "eval_runtime": 16.5084, | |
| "eval_samples_per_second": 605.753, | |
| "eval_steps_per_second": 18.96, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0002057142857142857, | |
| "loss": 1.6145, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00020376623376623372, | |
| "loss": 1.5818, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.0002018181818181818, | |
| "loss": 1.5862, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.41431556948798326, | |
| "eval_loss": 4.115560531616211, | |
| "eval_runtime": 16.3845, | |
| "eval_samples_per_second": 610.333, | |
| "eval_steps_per_second": 19.103, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00019987012987012986, | |
| "loss": 1.5816, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.0001979220779220779, | |
| "loss": 1.5614, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00019597402597402598, | |
| "loss": 1.5722, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_accuracy": 0.4119644723092999, | |
| "eval_loss": 4.146672248840332, | |
| "eval_runtime": 16.4614, | |
| "eval_samples_per_second": 607.482, | |
| "eval_steps_per_second": 19.014, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.000194025974025974, | |
| "loss": 1.5387, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00019207792207792207, | |
| "loss": 1.5353, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001901298701298701, | |
| "loss": 1.5182, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.414272030651341, | |
| "eval_loss": 4.154214382171631, | |
| "eval_runtime": 16.4446, | |
| "eval_samples_per_second": 608.101, | |
| "eval_steps_per_second": 19.034, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.0001881818181818182, | |
| "loss": 1.5075, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.00018623376623376622, | |
| "loss": 1.5203, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00018428571428571428, | |
| "loss": 1.5195, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.41157262277951934, | |
| "eval_loss": 4.271289825439453, | |
| "eval_runtime": 16.4758, | |
| "eval_samples_per_second": 606.95, | |
| "eval_steps_per_second": 18.998, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.0001823376623376623, | |
| "loss": 1.4973, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0001803896103896104, | |
| "loss": 1.6131, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.00017844155844155843, | |
| "loss": 1.3535, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_accuracy": 0.41322709857192613, | |
| "eval_loss": 4.2467217445373535, | |
| "eval_runtime": 16.3639, | |
| "eval_samples_per_second": 611.102, | |
| "eval_steps_per_second": 19.127, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00017649350649350648, | |
| "loss": 1.3275, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.00017454545454545452, | |
| "loss": 1.3251, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 0.0001725974025974026, | |
| "loss": 1.3322, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.40691396725879486, | |
| "eval_loss": 4.347814083099365, | |
| "eval_runtime": 16.4522, | |
| "eval_samples_per_second": 607.822, | |
| "eval_steps_per_second": 19.025, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 0.00017064935064935063, | |
| "loss": 1.3317, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0001687012987012987, | |
| "loss": 1.3386, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.00016675324675324672, | |
| "loss": 1.3253, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.40747997213514453, | |
| "eval_loss": 4.348999500274658, | |
| "eval_runtime": 16.4842, | |
| "eval_samples_per_second": 606.643, | |
| "eval_steps_per_second": 18.988, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0001648051948051948, | |
| "loss": 1.3325, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.00016285714285714284, | |
| "loss": 1.3262, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.0001609090909090909, | |
| "loss": 1.3052, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_accuracy": 0.41448972483455243, | |
| "eval_loss": 4.312870025634766, | |
| "eval_runtime": 16.4771, | |
| "eval_samples_per_second": 606.902, | |
| "eval_steps_per_second": 18.996, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.00015896103896103893, | |
| "loss": 1.3058, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.00015701298701298701, | |
| "loss": 1.297, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.00015506493506493505, | |
| "loss": 1.3093, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "eval_accuracy": 0.41527342389411354, | |
| "eval_loss": 4.299270153045654, | |
| "eval_runtime": 16.3482, | |
| "eval_samples_per_second": 611.689, | |
| "eval_steps_per_second": 19.146, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.0001531168831168831, | |
| "loss": 1.2818, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.00015116883116883114, | |
| "loss": 1.3043, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0001492207792207792, | |
| "loss": 1.2865, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_accuracy": 0.4116597004528039, | |
| "eval_loss": 4.3583269119262695, | |
| "eval_runtime": 17.7946, | |
| "eval_samples_per_second": 561.97, | |
| "eval_steps_per_second": 17.59, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.00014727272727272725, | |
| "loss": 1.2959, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0001453246753246753, | |
| "loss": 1.2998, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.00014337662337662337, | |
| "loss": 1.2934, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_accuracy": 0.4146203413444793, | |
| "eval_loss": 4.293191909790039, | |
| "eval_runtime": 16.4542, | |
| "eval_samples_per_second": 607.749, | |
| "eval_steps_per_second": 19.023, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.0001414285714285714, | |
| "loss": 1.2983, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00013948051948051946, | |
| "loss": 1.2949, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 0.00013753246753246752, | |
| "loss": 1.2967, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_accuracy": 0.4104841518634622, | |
| "eval_loss": 4.251837730407715, | |
| "eval_runtime": 16.3722, | |
| "eval_samples_per_second": 610.79, | |
| "eval_steps_per_second": 19.118, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.00013558441558441558, | |
| "loss": 1.2745, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 0.0001336363636363636, | |
| "loss": 1.276, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 0.00013168831168831167, | |
| "loss": 1.2706, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.4111807732497388, | |
| "eval_loss": 4.352819919586182, | |
| "eval_runtime": 16.376, | |
| "eval_samples_per_second": 610.648, | |
| "eval_steps_per_second": 19.113, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.00012974025974025972, | |
| "loss": 1.2837, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.00012779220779220778, | |
| "loss": 1.2779, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00012584415584415581, | |
| "loss": 1.2585, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.4151428073841867, | |
| "eval_loss": 4.317609786987305, | |
| "eval_runtime": 16.4566, | |
| "eval_samples_per_second": 607.659, | |
| "eval_steps_per_second": 19.02, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.0001238961038961039, | |
| "loss": 1.2726, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.00012194805194805194, | |
| "loss": 1.2667, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.3254, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_accuracy": 0.40573841866945315, | |
| "eval_loss": 4.5338358879089355, | |
| "eval_runtime": 16.961, | |
| "eval_samples_per_second": 589.587, | |
| "eval_steps_per_second": 18.454, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.00011805194805194805, | |
| "loss": 1.1489, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.00011610389610389609, | |
| "loss": 1.1305, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00011415584415584415, | |
| "loss": 1.1389, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.4110066179031696, | |
| "eval_loss": 4.393136978149414, | |
| "eval_runtime": 16.4067, | |
| "eval_samples_per_second": 609.508, | |
| "eval_steps_per_second": 19.078, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.0001122077922077922, | |
| "loss": 1.138, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 0.00011025974025974025, | |
| "loss": 1.1166, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 0.0001083116883116883, | |
| "loss": 1.149, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_accuracy": 0.4105712295367468, | |
| "eval_loss": 4.424084663391113, | |
| "eval_runtime": 16.3916, | |
| "eval_samples_per_second": 610.068, | |
| "eval_steps_per_second": 19.095, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.00010636363636363636, | |
| "loss": 1.1305, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.0001044155844155844, | |
| "loss": 1.1262, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.00010246753246753246, | |
| "loss": 1.1432, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "eval_accuracy": 0.41122431208638105, | |
| "eval_loss": 4.4350810050964355, | |
| "eval_runtime": 16.3705, | |
| "eval_samples_per_second": 610.854, | |
| "eval_steps_per_second": 19.12, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.0001005194805194805, | |
| "loss": 1.1285, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 9.857142857142856e-05, | |
| "loss": 1.1175, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 9.662337662337661e-05, | |
| "loss": 1.1383, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "eval_accuracy": 0.40865552072448624, | |
| "eval_loss": 4.405033111572266, | |
| "eval_runtime": 16.4548, | |
| "eval_samples_per_second": 607.724, | |
| "eval_steps_per_second": 19.022, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "step": 525, | |
| "total_flos": 3.4290279623915635e+17, | |
| "train_loss": 1.7241914413088844, | |
| "train_runtime": 16805.6947, | |
| "train_samples_per_second": 367.816, | |
| "train_steps_per_second": 0.046 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "eval_accuracy": 0.4117903169627308, | |
| "eval_loss": 4.301416397094727, | |
| "eval_runtime": 16.4623, | |
| "eval_samples_per_second": 607.449, | |
| "eval_steps_per_second": 19.013, | |
| "step": 525 | |
| } | |
| ], | |
| "max_steps": 770, | |
| "num_train_epochs": 5, | |
| "total_flos": 3.4290279623915635e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |