| { | |
| "best_metric": 0.9602224123182207, | |
| "best_model_checkpoint": "vit-msn-small-lateral_flow_ivalidation_green/checkpoint-32", | |
| "epoch": 92.3076923076923, | |
| "eval_steps": 500, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9230769230769231, | |
| "eval_accuracy": 0.9238665526090676, | |
| "eval_loss": 0.5034469366073608, | |
| "eval_runtime": 9.6118, | |
| "eval_samples_per_second": 243.243, | |
| "eval_steps_per_second": 3.849, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 13.861084938049316, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.704, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9409751924721984, | |
| "eval_loss": 0.45238828659057617, | |
| "eval_runtime": 9.8069, | |
| "eval_samples_per_second": 238.404, | |
| "eval_steps_per_second": 3.773, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "eval_accuracy": 0.9593669803250642, | |
| "eval_loss": 0.26334497332572937, | |
| "eval_runtime": 9.5224, | |
| "eval_samples_per_second": 245.527, | |
| "eval_steps_per_second": 3.886, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 5.300178527832031, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.505, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8092386655260907, | |
| "eval_loss": 0.4748455882072449, | |
| "eval_runtime": 9.8908, | |
| "eval_samples_per_second": 236.381, | |
| "eval_steps_per_second": 3.741, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 14.291367530822754, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4456, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "eval_accuracy": 0.9602224123182207, | |
| "eval_loss": 0.29173824191093445, | |
| "eval_runtime": 10.0407, | |
| "eval_samples_per_second": 232.852, | |
| "eval_steps_per_second": 3.685, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9221556886227545, | |
| "eval_loss": 0.2621428668498993, | |
| "eval_runtime": 9.9631, | |
| "eval_samples_per_second": 234.666, | |
| "eval_steps_per_second": 3.714, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 8.197896957397461, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3908, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "eval_accuracy": 0.8190761334473909, | |
| "eval_loss": 0.4519110918045044, | |
| "eval_runtime": 9.949, | |
| "eval_samples_per_second": 234.998, | |
| "eval_steps_per_second": 3.719, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 7.590042591094971, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.3628, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8622754491017964, | |
| "eval_loss": 0.4092731475830078, | |
| "eval_runtime": 10.096, | |
| "eval_samples_per_second": 231.578, | |
| "eval_steps_per_second": 3.665, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 8.923076923076923, | |
| "eval_accuracy": 0.935414884516681, | |
| "eval_loss": 0.2705248296260834, | |
| "eval_runtime": 10.1632, | |
| "eval_samples_per_second": 230.046, | |
| "eval_steps_per_second": 3.641, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 10.925556182861328, | |
| "learning_rate": 5e-05, | |
| "loss": 0.372, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8545765611633875, | |
| "eval_loss": 0.41367459297180176, | |
| "eval_runtime": 10.0011, | |
| "eval_samples_per_second": 233.774, | |
| "eval_steps_per_second": 3.7, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "grad_norm": 8.855840682983398, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 0.36, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.923076923076923, | |
| "eval_accuracy": 0.8815226689478186, | |
| "eval_loss": 0.34931161999702454, | |
| "eval_runtime": 9.9523, | |
| "eval_samples_per_second": 234.921, | |
| "eval_steps_per_second": 3.718, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9456800684345594, | |
| "eval_loss": 0.21901701390743256, | |
| "eval_runtime": 9.6913, | |
| "eval_samples_per_second": 241.248, | |
| "eval_steps_per_second": 3.818, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 7.141057014465332, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.36, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.923076923076923, | |
| "eval_accuracy": 0.9033361847733106, | |
| "eval_loss": 0.3190420866012573, | |
| "eval_runtime": 9.7798, | |
| "eval_samples_per_second": 239.064, | |
| "eval_steps_per_second": 3.783, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "grad_norm": 11.677677154541016, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3363, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.894781864841745, | |
| "eval_loss": 0.337951123714447, | |
| "eval_runtime": 9.7204, | |
| "eval_samples_per_second": 240.525, | |
| "eval_steps_per_second": 3.806, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 14.923076923076923, | |
| "eval_accuracy": 0.8982035928143712, | |
| "eval_loss": 0.3342379331588745, | |
| "eval_runtime": 9.9394, | |
| "eval_samples_per_second": 235.226, | |
| "eval_steps_per_second": 3.723, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 3.7460684776306152, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.327, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8327630453378957, | |
| "eval_loss": 0.4211990237236023, | |
| "eval_runtime": 9.8418, | |
| "eval_samples_per_second": 237.559, | |
| "eval_steps_per_second": 3.759, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "grad_norm": 12.34093952178955, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 0.3257, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "eval_accuracy": 0.7844311377245509, | |
| "eval_loss": 0.5167170763015747, | |
| "eval_runtime": 9.8234, | |
| "eval_samples_per_second": 238.004, | |
| "eval_steps_per_second": 3.767, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7275449101796407, | |
| "eval_loss": 0.5847879648208618, | |
| "eval_runtime": 9.927, | |
| "eval_samples_per_second": 235.518, | |
| "eval_steps_per_second": 3.727, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 18.46153846153846, | |
| "grad_norm": 5.46023416519165, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3175, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 18.923076923076923, | |
| "eval_accuracy": 0.8336184773310522, | |
| "eval_loss": 0.4090871810913086, | |
| "eval_runtime": 10.1427, | |
| "eval_samples_per_second": 230.511, | |
| "eval_steps_per_second": 3.648, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 9.724111557006836, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 0.3377, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9161676646706587, | |
| "eval_loss": 0.28380292654037476, | |
| "eval_runtime": 10.1341, | |
| "eval_samples_per_second": 230.706, | |
| "eval_steps_per_second": 3.651, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.923076923076923, | |
| "eval_accuracy": 0.7262617621899059, | |
| "eval_loss": 0.6106137633323669, | |
| "eval_runtime": 9.9506, | |
| "eval_samples_per_second": 234.962, | |
| "eval_steps_per_second": 3.718, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 21.53846153846154, | |
| "grad_norm": 5.13586950302124, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.3129, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7164242942686057, | |
| "eval_loss": 0.6294828057289124, | |
| "eval_runtime": 9.9909, | |
| "eval_samples_per_second": 234.013, | |
| "eval_steps_per_second": 3.703, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 22.923076923076923, | |
| "eval_accuracy": 0.5932420872540634, | |
| "eval_loss": 0.7897723913192749, | |
| "eval_runtime": 9.9526, | |
| "eval_samples_per_second": 234.912, | |
| "eval_steps_per_second": 3.718, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 23.076923076923077, | |
| "grad_norm": 15.414055824279785, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.3138, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.4846022241231822, | |
| "eval_loss": 0.9407968521118164, | |
| "eval_runtime": 9.7597, | |
| "eval_samples_per_second": 239.556, | |
| "eval_steps_per_second": 3.791, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 24.615384615384617, | |
| "grad_norm": 3.9247374534606934, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.3106, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 24.923076923076923, | |
| "eval_accuracy": 0.8832335329341318, | |
| "eval_loss": 0.34852102398872375, | |
| "eval_runtime": 9.8091, | |
| "eval_samples_per_second": 238.35, | |
| "eval_steps_per_second": 3.772, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7865697177074422, | |
| "eval_loss": 0.5201271176338196, | |
| "eval_runtime": 9.9386, | |
| "eval_samples_per_second": 235.245, | |
| "eval_steps_per_second": 3.723, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 26.153846153846153, | |
| "grad_norm": 5.763908386230469, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 0.3157, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 26.923076923076923, | |
| "eval_accuracy": 0.6672369546621043, | |
| "eval_loss": 0.72103351354599, | |
| "eval_runtime": 9.9351, | |
| "eval_samples_per_second": 235.327, | |
| "eval_steps_per_second": 3.724, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 27.692307692307693, | |
| "grad_norm": 5.021239757537842, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2896, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.6330196749358425, | |
| "eval_loss": 0.7980794906616211, | |
| "eval_runtime": 10.2592, | |
| "eval_samples_per_second": 227.892, | |
| "eval_steps_per_second": 3.607, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 28.923076923076923, | |
| "eval_accuracy": 0.6428571428571429, | |
| "eval_loss": 0.7667437791824341, | |
| "eval_runtime": 9.9249, | |
| "eval_samples_per_second": 235.57, | |
| "eval_steps_per_second": 3.728, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 29.23076923076923, | |
| "grad_norm": 8.59469985961914, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 0.2867, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.6544054747647562, | |
| "eval_loss": 0.7686835527420044, | |
| "eval_runtime": 10.0746, | |
| "eval_samples_per_second": 232.069, | |
| "eval_steps_per_second": 3.673, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "grad_norm": 5.601044654846191, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2786, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 30.923076923076923, | |
| "eval_accuracy": 0.5209580838323353, | |
| "eval_loss": 1.1714286804199219, | |
| "eval_runtime": 9.9004, | |
| "eval_samples_per_second": 236.151, | |
| "eval_steps_per_second": 3.737, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.42728828058169377, | |
| "eval_loss": 1.1744341850280762, | |
| "eval_runtime": 9.8292, | |
| "eval_samples_per_second": 237.862, | |
| "eval_steps_per_second": 3.764, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 32.30769230769231, | |
| "grad_norm": 4.0939507484436035, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.2823, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 32.92307692307692, | |
| "eval_accuracy": 0.5444824636441403, | |
| "eval_loss": 0.9260274767875671, | |
| "eval_runtime": 9.8098, | |
| "eval_samples_per_second": 238.334, | |
| "eval_steps_per_second": 3.772, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 33.84615384615385, | |
| "grad_norm": 5.409052848815918, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.2864, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.6920444824636441, | |
| "eval_loss": 0.7139692902565002, | |
| "eval_runtime": 9.9272, | |
| "eval_samples_per_second": 235.514, | |
| "eval_steps_per_second": 3.727, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 34.92307692307692, | |
| "eval_accuracy": 0.7331052181351583, | |
| "eval_loss": 0.6098384857177734, | |
| "eval_runtime": 9.8226, | |
| "eval_samples_per_second": 238.024, | |
| "eval_steps_per_second": 3.767, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 35.38461538461539, | |
| "grad_norm": 3.7521326541900635, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 0.2707, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6783575705731394, | |
| "eval_loss": 0.6992803812026978, | |
| "eval_runtime": 9.8614, | |
| "eval_samples_per_second": 237.087, | |
| "eval_steps_per_second": 3.752, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "grad_norm": 5.1208319664001465, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2921, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "eval_accuracy": 0.6176218990590248, | |
| "eval_loss": 0.8719092607498169, | |
| "eval_runtime": 10.0768, | |
| "eval_samples_per_second": 232.019, | |
| "eval_steps_per_second": 3.672, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.6060735671514115, | |
| "eval_loss": 0.8336823582649231, | |
| "eval_runtime": 9.8195, | |
| "eval_samples_per_second": 238.098, | |
| "eval_steps_per_second": 3.768, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 38.46153846153846, | |
| "grad_norm": 6.667710304260254, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.2849, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 38.92307692307692, | |
| "eval_accuracy": 0.825491873396065, | |
| "eval_loss": 0.4395623505115509, | |
| "eval_runtime": 9.6605, | |
| "eval_samples_per_second": 242.015, | |
| "eval_steps_per_second": 3.83, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 5.09765625, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.2657, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.501710863986313, | |
| "eval_loss": 1.0981603860855103, | |
| "eval_runtime": 9.9306, | |
| "eval_samples_per_second": 235.433, | |
| "eval_steps_per_second": 3.726, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 40.92307692307692, | |
| "eval_accuracy": 0.5175363558597091, | |
| "eval_loss": 1.093379259109497, | |
| "eval_runtime": 9.7012, | |
| "eval_samples_per_second": 241.001, | |
| "eval_steps_per_second": 3.814, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 41.53846153846154, | |
| "grad_norm": 3.8766767978668213, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.2659, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.636869118905047, | |
| "eval_loss": 0.8629169464111328, | |
| "eval_runtime": 9.9076, | |
| "eval_samples_per_second": 235.98, | |
| "eval_steps_per_second": 3.734, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 42.92307692307692, | |
| "eval_accuracy": 0.41402908468776733, | |
| "eval_loss": 1.4602264165878296, | |
| "eval_runtime": 9.7024, | |
| "eval_samples_per_second": 240.972, | |
| "eval_steps_per_second": 3.813, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 43.07692307692308, | |
| "grad_norm": 7.324178695678711, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.2645, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.3421727972626176, | |
| "eval_loss": 1.9095213413238525, | |
| "eval_runtime": 10.1083, | |
| "eval_samples_per_second": 231.294, | |
| "eval_steps_per_second": 3.66, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 44.61538461538461, | |
| "grad_norm": 3.6502673625946045, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.2424, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 44.92307692307692, | |
| "eval_accuracy": 0.43969204448246363, | |
| "eval_loss": 1.2180449962615967, | |
| "eval_runtime": 9.8511, | |
| "eval_samples_per_second": 237.334, | |
| "eval_steps_per_second": 3.756, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.6424294268605646, | |
| "eval_loss": 0.7686424255371094, | |
| "eval_runtime": 10.0098, | |
| "eval_samples_per_second": 233.572, | |
| "eval_steps_per_second": 3.696, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 46.15384615384615, | |
| "grad_norm": 4.265622138977051, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2495, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 46.92307692307692, | |
| "eval_accuracy": 0.5795551753635586, | |
| "eval_loss": 0.9899386763572693, | |
| "eval_runtime": 9.9941, | |
| "eval_samples_per_second": 233.939, | |
| "eval_steps_per_second": 3.702, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 47.69230769230769, | |
| "grad_norm": 2.909935474395752, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.2454, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.553464499572284, | |
| "eval_loss": 1.0290616750717163, | |
| "eval_runtime": 10.0955, | |
| "eval_samples_per_second": 231.587, | |
| "eval_steps_per_second": 3.665, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 48.92307692307692, | |
| "eval_accuracy": 0.6822070145423439, | |
| "eval_loss": 0.7534288167953491, | |
| "eval_runtime": 10.0774, | |
| "eval_samples_per_second": 232.004, | |
| "eval_steps_per_second": 3.672, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 49.23076923076923, | |
| "grad_norm": 6.560550212860107, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.2473, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.7091531223267751, | |
| "eval_loss": 0.6591421961784363, | |
| "eval_runtime": 10.0255, | |
| "eval_samples_per_second": 233.204, | |
| "eval_steps_per_second": 3.691, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 50.76923076923077, | |
| "grad_norm": 10.7844820022583, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2716, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 50.92307692307692, | |
| "eval_accuracy": 0.7455089820359282, | |
| "eval_loss": 0.58400559425354, | |
| "eval_runtime": 9.8847, | |
| "eval_samples_per_second": 236.527, | |
| "eval_steps_per_second": 3.743, | |
| "step": 331 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.47647562018819506, | |
| "eval_loss": 1.2430182695388794, | |
| "eval_runtime": 9.8852, | |
| "eval_samples_per_second": 236.515, | |
| "eval_steps_per_second": 3.743, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 52.30769230769231, | |
| "grad_norm": 7.81274938583374, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.234, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 52.92307692307692, | |
| "eval_accuracy": 0.5145423438836613, | |
| "eval_loss": 1.299268126487732, | |
| "eval_runtime": 10.1543, | |
| "eval_samples_per_second": 230.247, | |
| "eval_steps_per_second": 3.644, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 53.84615384615385, | |
| "grad_norm": 8.652073860168457, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.2482, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.7172797262617622, | |
| "eval_loss": 0.6042024493217468, | |
| "eval_runtime": 10.1638, | |
| "eval_samples_per_second": 230.033, | |
| "eval_steps_per_second": 3.64, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 54.92307692307692, | |
| "eval_accuracy": 0.6026518391787853, | |
| "eval_loss": 0.8891559839248657, | |
| "eval_runtime": 9.9638, | |
| "eval_samples_per_second": 234.65, | |
| "eval_steps_per_second": 3.713, | |
| "step": 357 | |
| }, | |
| { | |
| "epoch": 55.38461538461539, | |
| "grad_norm": 3.8207547664642334, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2339, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.316082121471343, | |
| "eval_loss": 1.8545583486557007, | |
| "eval_runtime": 9.9886, | |
| "eval_samples_per_second": 234.066, | |
| "eval_steps_per_second": 3.704, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 56.92307692307692, | |
| "grad_norm": 4.086548328399658, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.2461, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 56.92307692307692, | |
| "eval_accuracy": 0.5359281437125748, | |
| "eval_loss": 1.0858689546585083, | |
| "eval_runtime": 9.9977, | |
| "eval_samples_per_second": 233.854, | |
| "eval_steps_per_second": 3.701, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.6176218990590248, | |
| "eval_loss": 0.8690257668495178, | |
| "eval_runtime": 10.0183, | |
| "eval_samples_per_second": 233.374, | |
| "eval_steps_per_second": 3.693, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 58.46153846153846, | |
| "grad_norm": 4.676217079162598, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.2395, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 58.92307692307692, | |
| "eval_accuracy": 0.6693755346449958, | |
| "eval_loss": 0.7557449340820312, | |
| "eval_runtime": 9.868, | |
| "eval_samples_per_second": 236.928, | |
| "eval_steps_per_second": 3.75, | |
| "step": 383 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 7.573139667510986, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.2159, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.5701454234388366, | |
| "eval_loss": 1.053432822227478, | |
| "eval_runtime": 10.2535, | |
| "eval_samples_per_second": 228.02, | |
| "eval_steps_per_second": 3.609, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 60.92307692307692, | |
| "eval_accuracy": 0.5812660393498716, | |
| "eval_loss": 0.9855865240097046, | |
| "eval_runtime": 10.0776, | |
| "eval_samples_per_second": 231.999, | |
| "eval_steps_per_second": 3.672, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 61.53846153846154, | |
| "grad_norm": 4.489895820617676, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.2309, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.5500427715996579, | |
| "eval_loss": 0.9999585151672363, | |
| "eval_runtime": 9.7878, | |
| "eval_samples_per_second": 238.869, | |
| "eval_steps_per_second": 3.78, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 62.92307692307692, | |
| "eval_accuracy": 0.5179640718562875, | |
| "eval_loss": 1.1939594745635986, | |
| "eval_runtime": 9.8311, | |
| "eval_samples_per_second": 237.818, | |
| "eval_steps_per_second": 3.764, | |
| "step": 409 | |
| }, | |
| { | |
| "epoch": 63.07692307692308, | |
| "grad_norm": 6.18975830078125, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.2117, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.5153977758768178, | |
| "eval_loss": 1.1580592393875122, | |
| "eval_runtime": 10.0265, | |
| "eval_samples_per_second": 233.182, | |
| "eval_steps_per_second": 3.69, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 64.61538461538461, | |
| "grad_norm": 4.720950603485107, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.2307, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 64.92307692307692, | |
| "eval_accuracy": 0.5337895637296834, | |
| "eval_loss": 0.9987441897392273, | |
| "eval_runtime": 10.0605, | |
| "eval_samples_per_second": 232.395, | |
| "eval_steps_per_second": 3.678, | |
| "step": 422 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.5414884516680923, | |
| "eval_loss": 1.084990382194519, | |
| "eval_runtime": 9.9921, | |
| "eval_samples_per_second": 233.986, | |
| "eval_steps_per_second": 3.703, | |
| "step": 429 | |
| }, | |
| { | |
| "epoch": 66.15384615384616, | |
| "grad_norm": 5.033910274505615, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.2068, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 66.92307692307692, | |
| "eval_accuracy": 0.6013686911890505, | |
| "eval_loss": 0.942755401134491, | |
| "eval_runtime": 9.7471, | |
| "eval_samples_per_second": 239.866, | |
| "eval_steps_per_second": 3.796, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 67.6923076923077, | |
| "grad_norm": 5.34838342666626, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.2126, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.5115483319076134, | |
| "eval_loss": 1.237959861755371, | |
| "eval_runtime": 9.9382, | |
| "eval_samples_per_second": 235.253, | |
| "eval_steps_per_second": 3.723, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 68.92307692307692, | |
| "eval_accuracy": 0.5859709153122327, | |
| "eval_loss": 0.9992711544036865, | |
| "eval_runtime": 10.002, | |
| "eval_samples_per_second": 233.754, | |
| "eval_steps_per_second": 3.699, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 69.23076923076923, | |
| "grad_norm": 3.1523709297180176, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2176, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.5021385799828914, | |
| "eval_loss": 1.190958023071289, | |
| "eval_runtime": 9.7096, | |
| "eval_samples_per_second": 240.793, | |
| "eval_steps_per_second": 3.811, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 70.76923076923077, | |
| "grad_norm": 4.65359354019165, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.2096, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 70.92307692307692, | |
| "eval_accuracy": 0.5119760479041916, | |
| "eval_loss": 1.246795415878296, | |
| "eval_runtime": 9.658, | |
| "eval_samples_per_second": 242.079, | |
| "eval_steps_per_second": 3.831, | |
| "step": 461 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.6920444824636441, | |
| "eval_loss": 0.7588455677032471, | |
| "eval_runtime": 10.2262, | |
| "eval_samples_per_second": 228.628, | |
| "eval_steps_per_second": 3.618, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 72.3076923076923, | |
| "grad_norm": 4.450184345245361, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.2092, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 72.92307692307692, | |
| "eval_accuracy": 0.6308810949529512, | |
| "eval_loss": 0.900288999080658, | |
| "eval_runtime": 9.6458, | |
| "eval_samples_per_second": 242.386, | |
| "eval_steps_per_second": 3.836, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 73.84615384615384, | |
| "grad_norm": 4.230223178863525, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1968, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.564585115483319, | |
| "eval_loss": 1.1697088479995728, | |
| "eval_runtime": 9.7325, | |
| "eval_samples_per_second": 240.225, | |
| "eval_steps_per_second": 3.802, | |
| "step": 481 | |
| }, | |
| { | |
| "epoch": 74.92307692307692, | |
| "eval_accuracy": 0.6445680068434559, | |
| "eval_loss": 0.8789314031600952, | |
| "eval_runtime": 9.8672, | |
| "eval_samples_per_second": 236.946, | |
| "eval_steps_per_second": 3.75, | |
| "step": 487 | |
| }, | |
| { | |
| "epoch": 75.38461538461539, | |
| "grad_norm": 5.635646343231201, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.2027, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.5598802395209581, | |
| "eval_loss": 1.1352075338363647, | |
| "eval_runtime": 10.2152, | |
| "eval_samples_per_second": 228.875, | |
| "eval_steps_per_second": 3.622, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 76.92307692307692, | |
| "grad_norm": 4.811977386474609, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1965, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 76.92307692307692, | |
| "eval_accuracy": 0.5598802395209581, | |
| "eval_loss": 1.083630919456482, | |
| "eval_runtime": 9.8228, | |
| "eval_samples_per_second": 238.017, | |
| "eval_steps_per_second": 3.767, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.5902480752780154, | |
| "eval_loss": 1.018804669380188, | |
| "eval_runtime": 10.2662, | |
| "eval_samples_per_second": 227.739, | |
| "eval_steps_per_second": 3.604, | |
| "step": 507 | |
| }, | |
| { | |
| "epoch": 78.46153846153847, | |
| "grad_norm": 4.3275227546691895, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.2267, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 78.92307692307692, | |
| "eval_accuracy": 0.5975192472198461, | |
| "eval_loss": 1.0287189483642578, | |
| "eval_runtime": 10.2326, | |
| "eval_samples_per_second": 228.486, | |
| "eval_steps_per_second": 3.616, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 5.816257476806641, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.1967, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.6544054747647562, | |
| "eval_loss": 0.8465330004692078, | |
| "eval_runtime": 9.6333, | |
| "eval_samples_per_second": 242.7, | |
| "eval_steps_per_second": 3.841, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 80.92307692307692, | |
| "eval_accuracy": 0.5470487596236099, | |
| "eval_loss": 1.188087821006775, | |
| "eval_runtime": 9.8344, | |
| "eval_samples_per_second": 237.736, | |
| "eval_steps_per_second": 3.762, | |
| "step": 526 | |
| }, | |
| { | |
| "epoch": 81.53846153846153, | |
| "grad_norm": 4.023173809051514, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.1842, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.5367835757057314, | |
| "eval_loss": 1.235166072845459, | |
| "eval_runtime": 9.8519, | |
| "eval_samples_per_second": 237.315, | |
| "eval_steps_per_second": 3.756, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 82.92307692307692, | |
| "eval_accuracy": 0.5701454234388366, | |
| "eval_loss": 1.106431007385254, | |
| "eval_runtime": 10.0863, | |
| "eval_samples_per_second": 231.799, | |
| "eval_steps_per_second": 3.668, | |
| "step": 539 | |
| }, | |
| { | |
| "epoch": 83.07692307692308, | |
| "grad_norm": 5.945059299468994, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1952, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.6608212147134302, | |
| "eval_loss": 0.8087576031684875, | |
| "eval_runtime": 9.3282, | |
| "eval_samples_per_second": 250.637, | |
| "eval_steps_per_second": 3.966, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 84.61538461538461, | |
| "grad_norm": 4.164029121398926, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.1873, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 84.92307692307692, | |
| "eval_accuracy": 0.6086398631308811, | |
| "eval_loss": 0.9341749548912048, | |
| "eval_runtime": 9.6842, | |
| "eval_samples_per_second": 241.425, | |
| "eval_steps_per_second": 3.821, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.6056458511548332, | |
| "eval_loss": 0.9807350039482117, | |
| "eval_runtime": 9.6406, | |
| "eval_samples_per_second": 242.516, | |
| "eval_steps_per_second": 3.838, | |
| "step": 559 | |
| }, | |
| { | |
| "epoch": 86.15384615384616, | |
| "grad_norm": 5.312713146209717, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.185, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 86.92307692307692, | |
| "eval_accuracy": 0.5898203592814372, | |
| "eval_loss": 1.0164724588394165, | |
| "eval_runtime": 9.7581, | |
| "eval_samples_per_second": 239.596, | |
| "eval_steps_per_second": 3.792, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 87.6923076923077, | |
| "grad_norm": 5.478633403778076, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.1993, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.5474764756201882, | |
| "eval_loss": 1.1511483192443848, | |
| "eval_runtime": 9.905, | |
| "eval_samples_per_second": 236.042, | |
| "eval_steps_per_second": 3.735, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 88.92307692307692, | |
| "eval_accuracy": 0.5406330196749358, | |
| "eval_loss": 1.176562786102295, | |
| "eval_runtime": 10.1535, | |
| "eval_samples_per_second": 230.266, | |
| "eval_steps_per_second": 3.644, | |
| "step": 578 | |
| }, | |
| { | |
| "epoch": 89.23076923076923, | |
| "grad_norm": 3.2113797664642334, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.1707, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.5662959794696322, | |
| "eval_loss": 1.120088815689087, | |
| "eval_runtime": 9.7143, | |
| "eval_samples_per_second": 240.676, | |
| "eval_steps_per_second": 3.809, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 90.76923076923077, | |
| "grad_norm": 5.403631210327148, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.1852, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 90.92307692307692, | |
| "eval_accuracy": 0.5701454234388366, | |
| "eval_loss": 1.116164207458496, | |
| "eval_runtime": 10.0497, | |
| "eval_samples_per_second": 232.643, | |
| "eval_steps_per_second": 3.682, | |
| "step": 591 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.5680068434559452, | |
| "eval_loss": 1.1272914409637451, | |
| "eval_runtime": 9.9727, | |
| "eval_samples_per_second": 234.44, | |
| "eval_steps_per_second": 3.71, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "grad_norm": 4.397765159606934, | |
| "learning_rate": 0.0, | |
| "loss": 0.1904, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "eval_accuracy": 0.5680068434559452, | |
| "eval_loss": 1.1280397176742554, | |
| "eval_runtime": 10.1432, | |
| "eval_samples_per_second": 230.5, | |
| "eval_steps_per_second": 3.648, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "step": 600, | |
| "total_flos": 2.9138957540265e+18, | |
| "train_loss": 0.273075803120931, | |
| "train_runtime": 2186.3924, | |
| "train_samples_per_second": 73.774, | |
| "train_steps_per_second": 0.274 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.9138957540265e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |