| { | |
| "best_metric": 0.9772727272727273, | |
| "best_model_checkpoint": "beit-base-patch16-224-85-fold1/checkpoint-112", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5454545454545454, | |
| "eval_loss": 0.7308238744735718, | |
| "eval_runtime": 0.7278, | |
| "eval_samples_per_second": 60.457, | |
| "eval_steps_per_second": 2.748, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7927291393280029, | |
| "eval_runtime": 0.7032, | |
| "eval_samples_per_second": 62.571, | |
| "eval_steps_per_second": 2.844, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.9672332406044006, | |
| "eval_runtime": 0.6786, | |
| "eval_samples_per_second": 64.835, | |
| "eval_steps_per_second": 2.947, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6257070302963257, | |
| "eval_runtime": 0.6667, | |
| "eval_samples_per_second": 65.994, | |
| "eval_steps_per_second": 3.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 11.562158584594727, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6404, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.4645519554615021, | |
| "eval_runtime": 0.6715, | |
| "eval_samples_per_second": 65.525, | |
| "eval_steps_per_second": 2.978, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5648249983787537, | |
| "eval_runtime": 0.6749, | |
| "eval_samples_per_second": 65.19, | |
| "eval_steps_per_second": 2.963, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.4388914406299591, | |
| "eval_runtime": 0.6837, | |
| "eval_samples_per_second": 64.359, | |
| "eval_steps_per_second": 2.925, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.4522952139377594, | |
| "eval_runtime": 0.7027, | |
| "eval_samples_per_second": 62.616, | |
| "eval_steps_per_second": 2.846, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.46983444690704346, | |
| "eval_runtime": 0.6818, | |
| "eval_samples_per_second": 64.535, | |
| "eval_steps_per_second": 2.933, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 8.737435340881348, | |
| "learning_rate": 5e-05, | |
| "loss": 0.455, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3706974983215332, | |
| "eval_runtime": 0.6808, | |
| "eval_samples_per_second": 64.631, | |
| "eval_steps_per_second": 2.938, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.35938122868537903, | |
| "eval_runtime": 0.6781, | |
| "eval_samples_per_second": 64.883, | |
| "eval_steps_per_second": 2.949, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.6136276721954346, | |
| "eval_runtime": 0.6807, | |
| "eval_samples_per_second": 64.643, | |
| "eval_steps_per_second": 2.938, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.30222317576408386, | |
| "eval_runtime": 0.6745, | |
| "eval_samples_per_second": 65.234, | |
| "eval_steps_per_second": 2.965, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.29185205698013306, | |
| "eval_runtime": 0.6861, | |
| "eval_samples_per_second": 64.127, | |
| "eval_steps_per_second": 2.915, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 8.116207122802734, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3981, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3611550033092499, | |
| "eval_runtime": 0.6792, | |
| "eval_samples_per_second": 64.785, | |
| "eval_steps_per_second": 2.945, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.24922798573970795, | |
| "eval_runtime": 0.6808, | |
| "eval_samples_per_second": 64.63, | |
| "eval_steps_per_second": 2.938, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24598444998264313, | |
| "eval_runtime": 0.6781, | |
| "eval_samples_per_second": 64.888, | |
| "eval_steps_per_second": 2.949, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.2930772006511688, | |
| "eval_runtime": 0.6891, | |
| "eval_samples_per_second": 63.847, | |
| "eval_steps_per_second": 2.902, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.18223300576210022, | |
| "eval_runtime": 0.6759, | |
| "eval_samples_per_second": 65.101, | |
| "eval_steps_per_second": 2.959, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 15.090621948242188, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3257, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20600570738315582, | |
| "eval_runtime": 0.6756, | |
| "eval_samples_per_second": 65.128, | |
| "eval_steps_per_second": 2.96, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.21951809525489807, | |
| "eval_runtime": 0.6733, | |
| "eval_samples_per_second": 65.351, | |
| "eval_steps_per_second": 2.971, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2624146640300751, | |
| "eval_runtime": 0.6747, | |
| "eval_samples_per_second": 65.219, | |
| "eval_steps_per_second": 2.964, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2383754700422287, | |
| "eval_runtime": 0.6742, | |
| "eval_samples_per_second": 65.258, | |
| "eval_steps_per_second": 2.966, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1767176240682602, | |
| "eval_runtime": 0.6822, | |
| "eval_samples_per_second": 64.498, | |
| "eval_steps_per_second": 2.932, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 8.298179626464844, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2553, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20396804809570312, | |
| "eval_runtime": 0.6827, | |
| "eval_samples_per_second": 64.45, | |
| "eval_steps_per_second": 2.93, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1980859339237213, | |
| "eval_runtime": 0.6738, | |
| "eval_samples_per_second": 65.297, | |
| "eval_steps_per_second": 2.968, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.18346600234508514, | |
| "eval_runtime": 0.6738, | |
| "eval_samples_per_second": 65.303, | |
| "eval_steps_per_second": 2.968, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1820111870765686, | |
| "eval_runtime": 0.6748, | |
| "eval_samples_per_second": 65.209, | |
| "eval_steps_per_second": 2.964, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.1465969681739807, | |
| "eval_runtime": 0.6762, | |
| "eval_samples_per_second": 65.067, | |
| "eval_steps_per_second": 2.958, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 6.1258087158203125, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2083, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.16679802536964417, | |
| "eval_runtime": 0.6722, | |
| "eval_samples_per_second": 65.452, | |
| "eval_steps_per_second": 2.975, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2228713482618332, | |
| "eval_runtime": 0.6833, | |
| "eval_samples_per_second": 64.389, | |
| "eval_steps_per_second": 2.927, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.17827552556991577, | |
| "eval_runtime": 0.6819, | |
| "eval_samples_per_second": 64.521, | |
| "eval_steps_per_second": 2.933, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.19438548386096954, | |
| "eval_runtime": 0.6804, | |
| "eval_samples_per_second": 64.672, | |
| "eval_steps_per_second": 2.94, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.30252915620803833, | |
| "eval_runtime": 0.6694, | |
| "eval_samples_per_second": 65.735, | |
| "eval_steps_per_second": 2.988, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 10.506503105163574, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.2353, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4457201659679413, | |
| "eval_runtime": 0.6705, | |
| "eval_samples_per_second": 65.624, | |
| "eval_steps_per_second": 2.983, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.27586841583251953, | |
| "eval_runtime": 0.7004, | |
| "eval_samples_per_second": 62.824, | |
| "eval_steps_per_second": 2.856, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2178523689508438, | |
| "eval_runtime": 0.705, | |
| "eval_samples_per_second": 62.41, | |
| "eval_steps_per_second": 2.837, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.39105215668678284, | |
| "eval_runtime": 0.7066, | |
| "eval_samples_per_second": 62.271, | |
| "eval_steps_per_second": 2.831, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.5785248279571533, | |
| "eval_runtime": 0.7084, | |
| "eval_samples_per_second": 62.115, | |
| "eval_steps_per_second": 2.823, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 3.042407751083374, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1782, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2338746190071106, | |
| "eval_runtime": 0.6767, | |
| "eval_samples_per_second": 65.017, | |
| "eval_steps_per_second": 2.955, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.23022708296775818, | |
| "eval_runtime": 0.6612, | |
| "eval_samples_per_second": 66.541, | |
| "eval_steps_per_second": 3.025, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.39668968319892883, | |
| "eval_runtime": 0.6712, | |
| "eval_samples_per_second": 65.555, | |
| "eval_steps_per_second": 2.98, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.4447011649608612, | |
| "eval_runtime": 0.6775, | |
| "eval_samples_per_second": 64.94, | |
| "eval_steps_per_second": 2.952, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20197413861751556, | |
| "eval_runtime": 0.6933, | |
| "eval_samples_per_second": 63.464, | |
| "eval_steps_per_second": 2.885, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 6.790353775024414, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.2059, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.191122904419899, | |
| "eval_runtime": 0.6988, | |
| "eval_samples_per_second": 62.967, | |
| "eval_steps_per_second": 2.862, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.26089540123939514, | |
| "eval_runtime": 0.6777, | |
| "eval_samples_per_second": 64.925, | |
| "eval_steps_per_second": 2.951, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2924693822860718, | |
| "eval_runtime": 0.6842, | |
| "eval_samples_per_second": 64.304, | |
| "eval_steps_per_second": 2.923, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2078903764486313, | |
| "eval_runtime": 0.693, | |
| "eval_samples_per_second": 63.488, | |
| "eval_steps_per_second": 2.886, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.18529368937015533, | |
| "eval_runtime": 0.6921, | |
| "eval_samples_per_second": 63.572, | |
| "eval_steps_per_second": 2.89, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 4.374385833740234, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1706, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2860322594642639, | |
| "eval_runtime": 0.697, | |
| "eval_samples_per_second": 63.13, | |
| "eval_steps_per_second": 2.87, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3734706938266754, | |
| "eval_runtime": 0.6878, | |
| "eval_samples_per_second": 63.971, | |
| "eval_steps_per_second": 2.908, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19677944481372833, | |
| "eval_runtime": 0.6875, | |
| "eval_samples_per_second": 63.996, | |
| "eval_steps_per_second": 2.909, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.17221951484680176, | |
| "eval_runtime": 0.679, | |
| "eval_samples_per_second": 64.798, | |
| "eval_steps_per_second": 2.945, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.312281996011734, | |
| "eval_runtime": 0.6832, | |
| "eval_samples_per_second": 64.399, | |
| "eval_steps_per_second": 2.927, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 3.807600259780884, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1429, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.32969653606414795, | |
| "eval_runtime": 0.6845, | |
| "eval_samples_per_second": 64.284, | |
| "eval_steps_per_second": 2.922, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9772727272727273, | |
| "eval_loss": 0.14295423030853271, | |
| "eval_runtime": 0.6744, | |
| "eval_samples_per_second": 65.244, | |
| "eval_steps_per_second": 2.966, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.9772727272727273, | |
| "eval_loss": 0.11342933028936386, | |
| "eval_runtime": 0.6865, | |
| "eval_samples_per_second": 64.09, | |
| "eval_steps_per_second": 2.913, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.231190487742424, | |
| "eval_runtime": 0.6772, | |
| "eval_samples_per_second": 64.97, | |
| "eval_steps_per_second": 2.953, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.28261473774909973, | |
| "eval_runtime": 0.6857, | |
| "eval_samples_per_second": 64.163, | |
| "eval_steps_per_second": 2.917, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 3.425337314605713, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1325, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2417471557855606, | |
| "eval_runtime": 0.6829, | |
| "eval_samples_per_second": 64.43, | |
| "eval_steps_per_second": 2.929, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.139329731464386, | |
| "eval_runtime": 0.6819, | |
| "eval_samples_per_second": 64.521, | |
| "eval_steps_per_second": 2.933, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.21779733896255493, | |
| "eval_runtime": 0.6874, | |
| "eval_samples_per_second": 64.008, | |
| "eval_steps_per_second": 2.909, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3990946114063263, | |
| "eval_runtime": 0.6762, | |
| "eval_samples_per_second": 65.068, | |
| "eval_steps_per_second": 2.958, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.33248183131217957, | |
| "eval_runtime": 0.6851, | |
| "eval_samples_per_second": 64.222, | |
| "eval_steps_per_second": 2.919, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 3.3198602199554443, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.1481, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.23272232711315155, | |
| "eval_runtime": 0.6764, | |
| "eval_samples_per_second": 65.054, | |
| "eval_steps_per_second": 2.957, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.28850874304771423, | |
| "eval_runtime": 0.6864, | |
| "eval_samples_per_second": 64.098, | |
| "eval_steps_per_second": 2.914, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.3575742840766907, | |
| "eval_runtime": 0.692, | |
| "eval_samples_per_second": 63.587, | |
| "eval_steps_per_second": 2.89, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.26858440041542053, | |
| "eval_runtime": 0.6743, | |
| "eval_samples_per_second": 65.257, | |
| "eval_steps_per_second": 2.966, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.1716589480638504, | |
| "eval_runtime": 0.6829, | |
| "eval_samples_per_second": 64.43, | |
| "eval_steps_per_second": 2.929, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 4.273709774017334, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1237, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.14934584498405457, | |
| "eval_runtime": 0.6712, | |
| "eval_samples_per_second": 65.551, | |
| "eval_steps_per_second": 2.98, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1428576558828354, | |
| "eval_runtime": 0.6789, | |
| "eval_samples_per_second": 64.814, | |
| "eval_steps_per_second": 2.946, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1789616346359253, | |
| "eval_runtime": 0.6831, | |
| "eval_samples_per_second": 64.412, | |
| "eval_steps_per_second": 2.928, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1590358167886734, | |
| "eval_runtime": 0.6785, | |
| "eval_samples_per_second": 64.853, | |
| "eval_steps_per_second": 2.948, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.19707775115966797, | |
| "eval_runtime": 0.6779, | |
| "eval_samples_per_second": 64.91, | |
| "eval_steps_per_second": 2.95, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 3.867281436920166, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.105, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.22291280329227448, | |
| "eval_runtime": 0.6754, | |
| "eval_samples_per_second": 65.149, | |
| "eval_steps_per_second": 2.961, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.17888131737709045, | |
| "eval_runtime": 0.6771, | |
| "eval_samples_per_second": 64.982, | |
| "eval_steps_per_second": 2.954, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.16707843542099, | |
| "eval_runtime": 0.6843, | |
| "eval_samples_per_second": 64.299, | |
| "eval_steps_per_second": 2.923, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.24351829290390015, | |
| "eval_runtime": 0.6914, | |
| "eval_samples_per_second": 63.642, | |
| "eval_steps_per_second": 2.893, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2658483386039734, | |
| "eval_runtime": 0.6856, | |
| "eval_samples_per_second": 64.176, | |
| "eval_steps_per_second": 2.917, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 2.501020908355713, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0923, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20919661223888397, | |
| "eval_runtime": 0.6861, | |
| "eval_samples_per_second": 64.132, | |
| "eval_steps_per_second": 2.915, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.17477592825889587, | |
| "eval_runtime": 0.6775, | |
| "eval_samples_per_second": 64.946, | |
| "eval_steps_per_second": 2.952, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1727462261915207, | |
| "eval_runtime": 0.6757, | |
| "eval_samples_per_second": 65.116, | |
| "eval_steps_per_second": 2.96, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1945178508758545, | |
| "eval_runtime": 0.6723, | |
| "eval_samples_per_second": 65.443, | |
| "eval_steps_per_second": 2.975, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.24294686317443848, | |
| "eval_runtime": 0.6772, | |
| "eval_samples_per_second": 64.978, | |
| "eval_steps_per_second": 2.954, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 3.9196958541870117, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.1033, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2795504331588745, | |
| "eval_runtime": 0.6951, | |
| "eval_samples_per_second": 63.299, | |
| "eval_steps_per_second": 2.877, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2548198401927948, | |
| "eval_runtime": 0.6876, | |
| "eval_samples_per_second": 63.99, | |
| "eval_steps_per_second": 2.909, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2379317283630371, | |
| "eval_runtime": 0.6842, | |
| "eval_samples_per_second": 64.312, | |
| "eval_steps_per_second": 2.923, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2408948391675949, | |
| "eval_runtime": 0.6803, | |
| "eval_samples_per_second": 64.68, | |
| "eval_steps_per_second": 2.94, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24209485948085785, | |
| "eval_runtime": 0.6754, | |
| "eval_samples_per_second": 65.142, | |
| "eval_steps_per_second": 2.961, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 2.0967917442321777, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1073, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.23317375779151917, | |
| "eval_runtime": 0.6747, | |
| "eval_samples_per_second": 65.214, | |
| "eval_steps_per_second": 2.964, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.22311018407344818, | |
| "eval_runtime": 0.6758, | |
| "eval_samples_per_second": 65.111, | |
| "eval_steps_per_second": 2.96, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.21525071561336517, | |
| "eval_runtime": 0.6873, | |
| "eval_samples_per_second": 64.018, | |
| "eval_steps_per_second": 2.91, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20876212418079376, | |
| "eval_runtime": 0.6844, | |
| "eval_samples_per_second": 64.295, | |
| "eval_steps_per_second": 2.922, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20578283071517944, | |
| "eval_runtime": 0.6983, | |
| "eval_samples_per_second": 63.011, | |
| "eval_steps_per_second": 2.864, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 2.9749207496643066, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.104, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20398248732089996, | |
| "eval_runtime": 0.6745, | |
| "eval_samples_per_second": 65.237, | |
| "eval_steps_per_second": 2.965, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20461073517799377, | |
| "eval_runtime": 0.673, | |
| "eval_samples_per_second": 65.381, | |
| "eval_steps_per_second": 2.972, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20431600511074066, | |
| "eval_runtime": 0.6746, | |
| "eval_samples_per_second": 65.228, | |
| "eval_steps_per_second": 2.965, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20564264059066772, | |
| "eval_runtime": 0.6795, | |
| "eval_samples_per_second": 64.751, | |
| "eval_steps_per_second": 2.943, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20811475813388824, | |
| "eval_runtime": 0.6891, | |
| "eval_samples_per_second": 63.849, | |
| "eval_steps_per_second": 2.902, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 2.3249521255493164, | |
| "learning_rate": 0.0, | |
| "loss": 0.0896, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2097294181585312, | |
| "eval_runtime": 0.686, | |
| "eval_samples_per_second": 64.143, | |
| "eval_steps_per_second": 2.916, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.8822505078960128e+18, | |
| "train_loss": 0.21107831209897995, | |
| "train_runtime": 1239.3905, | |
| "train_samples_per_second": 19.606, | |
| "train_steps_per_second": 0.161 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9772727272727273, | |
| "eval_loss": 0.14295423030853271, | |
| "eval_runtime": 0.7189, | |
| "eval_samples_per_second": 61.203, | |
| "eval_steps_per_second": 2.782, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 1.8822505078960128e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |