| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 76.8, | |
| "eval_steps": 500, | |
| "global_step": 480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.22988505747126436, | |
| "eval_loss": 1.7957335710525513, | |
| "eval_runtime": 2.3499, | |
| "eval_samples_per_second": 37.022, | |
| "eval_steps_per_second": 1.277, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 5.402665615081787, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 1.8656, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_accuracy": 0.27586206896551724, | |
| "eval_loss": 1.7703895568847656, | |
| "eval_runtime": 2.2621, | |
| "eval_samples_per_second": 38.46, | |
| "eval_steps_per_second": 1.326, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_accuracy": 0.3218390804597701, | |
| "eval_loss": 1.738166332244873, | |
| "eval_runtime": 2.047, | |
| "eval_samples_per_second": 42.502, | |
| "eval_steps_per_second": 1.466, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 9.074636459350586, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.7835, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3793103448275862, | |
| "eval_loss": 1.6673917770385742, | |
| "eval_runtime": 2.0721, | |
| "eval_samples_per_second": 41.987, | |
| "eval_steps_per_second": 1.448, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 23.296255111694336, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.664, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.42528735632183906, | |
| "eval_loss": 1.5981522798538208, | |
| "eval_runtime": 2.0749, | |
| "eval_samples_per_second": 41.93, | |
| "eval_steps_per_second": 1.446, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "eval_accuracy": 0.4367816091954023, | |
| "eval_loss": 1.4861106872558594, | |
| "eval_runtime": 2.0842, | |
| "eval_samples_per_second": 41.743, | |
| "eval_steps_per_second": 1.439, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 36.56019592285156, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.5072, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "eval_accuracy": 0.47126436781609193, | |
| "eval_loss": 1.3644713163375854, | |
| "eval_runtime": 2.0344, | |
| "eval_samples_per_second": 42.765, | |
| "eval_steps_per_second": 1.475, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 70.99100494384766, | |
| "learning_rate": 4.976851851851852e-05, | |
| "loss": 1.3304, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.45977011494252873, | |
| "eval_loss": 1.285918116569519, | |
| "eval_runtime": 2.0671, | |
| "eval_samples_per_second": 42.088, | |
| "eval_steps_per_second": 1.451, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_accuracy": 0.47126436781609193, | |
| "eval_loss": 1.2795610427856445, | |
| "eval_runtime": 2.0462, | |
| "eval_samples_per_second": 42.519, | |
| "eval_steps_per_second": 1.466, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 60.155181884765625, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.1651, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "eval_accuracy": 0.5172413793103449, | |
| "eval_loss": 1.2455964088439941, | |
| "eval_runtime": 2.0479, | |
| "eval_samples_per_second": 42.483, | |
| "eval_steps_per_second": 1.465, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "eval_accuracy": 0.5402298850574713, | |
| "eval_loss": 1.1666686534881592, | |
| "eval_runtime": 2.0486, | |
| "eval_samples_per_second": 42.468, | |
| "eval_steps_per_second": 1.464, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "grad_norm": 17.20172119140625, | |
| "learning_rate": 4.745370370370371e-05, | |
| "loss": 1.0876, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5632183908045977, | |
| "eval_loss": 1.1510032415390015, | |
| "eval_runtime": 2.0486, | |
| "eval_samples_per_second": 42.468, | |
| "eval_steps_per_second": 1.464, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "grad_norm": 98.55331420898438, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.0046, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_accuracy": 0.6091954022988506, | |
| "eval_loss": 1.0509852170944214, | |
| "eval_runtime": 2.2, | |
| "eval_samples_per_second": 39.546, | |
| "eval_steps_per_second": 1.364, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 13.92, | |
| "eval_accuracy": 0.5862068965517241, | |
| "eval_loss": 1.033838152885437, | |
| "eval_runtime": 2.0133, | |
| "eval_samples_per_second": 43.212, | |
| "eval_steps_per_second": 1.49, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "grad_norm": 53.443302154541016, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 0.9465, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "eval_accuracy": 0.5862068965517241, | |
| "eval_loss": 0.9883113503456116, | |
| "eval_runtime": 2.05, | |
| "eval_samples_per_second": 42.439, | |
| "eval_steps_per_second": 1.463, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 30.475088119506836, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 0.8699, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5632183908045977, | |
| "eval_loss": 0.9881502389907837, | |
| "eval_runtime": 2.0881, | |
| "eval_samples_per_second": 41.664, | |
| "eval_steps_per_second": 1.437, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "eval_accuracy": 0.5747126436781609, | |
| "eval_loss": 0.9276102781295776, | |
| "eval_runtime": 2.0889, | |
| "eval_samples_per_second": 41.648, | |
| "eval_steps_per_second": 1.436, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "grad_norm": 21.802074432373047, | |
| "learning_rate": 4.282407407407408e-05, | |
| "loss": 0.7969, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "eval_accuracy": 0.5862068965517241, | |
| "eval_loss": 0.9144545197486877, | |
| "eval_runtime": 2.0314, | |
| "eval_samples_per_second": 42.828, | |
| "eval_steps_per_second": 1.477, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.8143898844718933, | |
| "eval_runtime": 2.0134, | |
| "eval_samples_per_second": 43.21, | |
| "eval_steps_per_second": 1.49, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "grad_norm": 58.785552978515625, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.7254, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.7586901187896729, | |
| "eval_runtime": 2.0526, | |
| "eval_samples_per_second": 42.386, | |
| "eval_steps_per_second": 1.462, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "grad_norm": 24.079566955566406, | |
| "learning_rate": 4.0509259259259265e-05, | |
| "loss": 0.6447, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "eval_accuracy": 0.7471264367816092, | |
| "eval_loss": 0.6990374326705933, | |
| "eval_runtime": 2.0625, | |
| "eval_samples_per_second": 42.182, | |
| "eval_steps_per_second": 1.455, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.7041503190994263, | |
| "eval_runtime": 2.0267, | |
| "eval_samples_per_second": 42.926, | |
| "eval_steps_per_second": 1.48, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "grad_norm": 24.671459197998047, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 0.6021, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 22.88, | |
| "eval_accuracy": 0.7701149425287356, | |
| "eval_loss": 0.6526182293891907, | |
| "eval_runtime": 2.1122, | |
| "eval_samples_per_second": 41.189, | |
| "eval_steps_per_second": 1.42, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 55.466121673583984, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.516, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.6485260128974915, | |
| "eval_runtime": 2.0692, | |
| "eval_samples_per_second": 42.046, | |
| "eval_steps_per_second": 1.45, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.5802629590034485, | |
| "eval_runtime": 2.0421, | |
| "eval_samples_per_second": 42.603, | |
| "eval_steps_per_second": 1.469, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "grad_norm": 17.66895294189453, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.4497, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 25.92, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.6084781289100647, | |
| "eval_runtime": 2.0191, | |
| "eval_samples_per_second": 43.088, | |
| "eval_steps_per_second": 1.486, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.6094852685928345, | |
| "eval_runtime": 1.9897, | |
| "eval_samples_per_second": 43.724, | |
| "eval_steps_per_second": 1.508, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "grad_norm": 31.39649200439453, | |
| "learning_rate": 3.587962962962963e-05, | |
| "loss": 0.3935, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.5372287034988403, | |
| "eval_runtime": 2.0637, | |
| "eval_samples_per_second": 42.158, | |
| "eval_steps_per_second": 1.454, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "grad_norm": 31.86827278137207, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.3321, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.5828755497932434, | |
| "eval_runtime": 2.1428, | |
| "eval_samples_per_second": 40.6, | |
| "eval_steps_per_second": 1.4, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 29.92, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.6204901337623596, | |
| "eval_runtime": 2.0154, | |
| "eval_samples_per_second": 43.168, | |
| "eval_steps_per_second": 1.489, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "grad_norm": 42.88612747192383, | |
| "learning_rate": 3.3564814814814815e-05, | |
| "loss": 0.3007, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 30.88, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.5149825811386108, | |
| "eval_runtime": 2.0492, | |
| "eval_samples_per_second": 42.456, | |
| "eval_steps_per_second": 1.464, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 30.13237190246582, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.2618, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8390804597701149, | |
| "eval_loss": 0.6068965196609497, | |
| "eval_runtime": 2.0657, | |
| "eval_samples_per_second": 42.117, | |
| "eval_steps_per_second": 1.452, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "eval_accuracy": 0.8390804597701149, | |
| "eval_loss": 0.5272508859634399, | |
| "eval_runtime": 2.0395, | |
| "eval_samples_per_second": 42.657, | |
| "eval_steps_per_second": 1.471, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 33.6, | |
| "grad_norm": 24.97075080871582, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.2411, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 33.92, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.4726714789867401, | |
| "eval_runtime": 2.0758, | |
| "eval_samples_per_second": 41.912, | |
| "eval_steps_per_second": 1.445, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 34.88, | |
| "eval_accuracy": 0.8735632183908046, | |
| "eval_loss": 0.4611084461212158, | |
| "eval_runtime": 2.0264, | |
| "eval_samples_per_second": 42.934, | |
| "eval_steps_per_second": 1.48, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "grad_norm": 60.3193359375, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 0.2108, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5696073770523071, | |
| "eval_runtime": 2.0919, | |
| "eval_samples_per_second": 41.589, | |
| "eval_steps_per_second": 1.434, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "grad_norm": 16.915546417236328, | |
| "learning_rate": 2.8935185185185186e-05, | |
| "loss": 0.2143, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 36.96, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.49439194798469543, | |
| "eval_runtime": 2.0923, | |
| "eval_samples_per_second": 41.58, | |
| "eval_steps_per_second": 1.434, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 37.92, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.5627816915512085, | |
| "eval_runtime": 2.0503, | |
| "eval_samples_per_second": 42.432, | |
| "eval_steps_per_second": 1.463, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 38.4, | |
| "grad_norm": 14.699493408203125, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1663, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 38.88, | |
| "eval_accuracy": 0.8045977011494253, | |
| "eval_loss": 0.6131365895271301, | |
| "eval_runtime": 2.0693, | |
| "eval_samples_per_second": 42.044, | |
| "eval_steps_per_second": 1.45, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 25.7874755859375, | |
| "learning_rate": 2.6620370370370372e-05, | |
| "loss": 0.1714, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.4961901605129242, | |
| "eval_runtime": 2.0252, | |
| "eval_samples_per_second": 42.959, | |
| "eval_steps_per_second": 1.481, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 40.96, | |
| "eval_accuracy": 0.8390804597701149, | |
| "eval_loss": 0.5022612810134888, | |
| "eval_runtime": 2.127, | |
| "eval_samples_per_second": 40.904, | |
| "eval_steps_per_second": 1.41, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 41.6, | |
| "grad_norm": 24.087005615234375, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.174, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 41.92, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.48418501019477844, | |
| "eval_runtime": 2.0168, | |
| "eval_samples_per_second": 43.137, | |
| "eval_steps_per_second": 1.487, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 42.88, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.46790340542793274, | |
| "eval_runtime": 2.0909, | |
| "eval_samples_per_second": 41.609, | |
| "eval_steps_per_second": 1.435, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 43.2, | |
| "grad_norm": 13.284588813781738, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.138, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.6270841956138611, | |
| "eval_runtime": 2.1069, | |
| "eval_samples_per_second": 41.294, | |
| "eval_steps_per_second": 1.424, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "grad_norm": 14.41830062866211, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.1437, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 44.96, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5325595736503601, | |
| "eval_runtime": 2.1982, | |
| "eval_samples_per_second": 39.578, | |
| "eval_steps_per_second": 1.365, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 45.92, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.5655315518379211, | |
| "eval_runtime": 2.0683, | |
| "eval_samples_per_second": 42.063, | |
| "eval_steps_per_second": 1.45, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 46.4, | |
| "grad_norm": 17.588279724121094, | |
| "learning_rate": 2.1990740740740743e-05, | |
| "loss": 0.136, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 46.88, | |
| "eval_accuracy": 0.8390804597701149, | |
| "eval_loss": 0.46718767285346985, | |
| "eval_runtime": 2.0892, | |
| "eval_samples_per_second": 41.643, | |
| "eval_steps_per_second": 1.436, | |
| "step": 293 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 22.864524841308594, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.1401, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.498960942029953, | |
| "eval_runtime": 2.0484, | |
| "eval_samples_per_second": 42.471, | |
| "eval_steps_per_second": 1.465, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 48.96, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.5445386171340942, | |
| "eval_runtime": 2.0365, | |
| "eval_samples_per_second": 42.721, | |
| "eval_steps_per_second": 1.473, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 49.6, | |
| "grad_norm": 22.651620864868164, | |
| "learning_rate": 1.967592592592593e-05, | |
| "loss": 0.1281, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 49.92, | |
| "eval_accuracy": 0.8735632183908046, | |
| "eval_loss": 0.47610902786254883, | |
| "eval_runtime": 2.1166, | |
| "eval_samples_per_second": 41.104, | |
| "eval_steps_per_second": 1.417, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 50.88, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5665103793144226, | |
| "eval_runtime": 2.1168, | |
| "eval_samples_per_second": 41.1, | |
| "eval_steps_per_second": 1.417, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 51.2, | |
| "grad_norm": 26.539594650268555, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1156, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5089926719665527, | |
| "eval_runtime": 2.0775, | |
| "eval_samples_per_second": 41.877, | |
| "eval_steps_per_second": 1.444, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 52.8, | |
| "grad_norm": 23.464221954345703, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.0981, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 52.96, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5152259469032288, | |
| "eval_runtime": 2.0607, | |
| "eval_samples_per_second": 42.219, | |
| "eval_steps_per_second": 1.456, | |
| "step": 331 | |
| }, | |
| { | |
| "epoch": 53.92, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.5466004610061646, | |
| "eval_runtime": 2.0591, | |
| "eval_samples_per_second": 42.251, | |
| "eval_steps_per_second": 1.457, | |
| "step": 337 | |
| }, | |
| { | |
| "epoch": 54.4, | |
| "grad_norm": 14.581974983215332, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.1055, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 54.88, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.5390048623085022, | |
| "eval_runtime": 2.0443, | |
| "eval_samples_per_second": 42.558, | |
| "eval_steps_per_second": 1.468, | |
| "step": 343 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "grad_norm": 14.774139404296875, | |
| "learning_rate": 1.5046296296296297e-05, | |
| "loss": 0.112, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5574498176574707, | |
| "eval_runtime": 2.0874, | |
| "eval_samples_per_second": 41.679, | |
| "eval_steps_per_second": 1.437, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 56.96, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5448784828186035, | |
| "eval_runtime": 2.0514, | |
| "eval_samples_per_second": 42.41, | |
| "eval_steps_per_second": 1.462, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 57.6, | |
| "grad_norm": 18.17756462097168, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0855, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 57.92, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5390240550041199, | |
| "eval_runtime": 2.077, | |
| "eval_samples_per_second": 41.888, | |
| "eval_steps_per_second": 1.444, | |
| "step": 362 | |
| }, | |
| { | |
| "epoch": 58.88, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5206344723701477, | |
| "eval_runtime": 2.0568, | |
| "eval_samples_per_second": 42.299, | |
| "eval_steps_per_second": 1.459, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 59.2, | |
| "grad_norm": 40.29678726196289, | |
| "learning_rate": 1.2731481481481482e-05, | |
| "loss": 0.0899, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.5475941300392151, | |
| "eval_runtime": 2.063, | |
| "eval_samples_per_second": 42.172, | |
| "eval_steps_per_second": 1.454, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 60.8, | |
| "grad_norm": 33.910377502441406, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.1026, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 60.96, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5344437956809998, | |
| "eval_runtime": 2.298, | |
| "eval_samples_per_second": 37.858, | |
| "eval_steps_per_second": 1.305, | |
| "step": 381 | |
| }, | |
| { | |
| "epoch": 61.92, | |
| "eval_accuracy": 0.8390804597701149, | |
| "eval_loss": 0.553070068359375, | |
| "eval_runtime": 2.1032, | |
| "eval_samples_per_second": 41.366, | |
| "eval_steps_per_second": 1.426, | |
| "step": 387 | |
| }, | |
| { | |
| "epoch": 62.4, | |
| "grad_norm": 13.71580982208252, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.0799, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 62.88, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.57228684425354, | |
| "eval_runtime": 2.0779, | |
| "eval_samples_per_second": 41.868, | |
| "eval_steps_per_second": 1.444, | |
| "step": 393 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "grad_norm": 28.238468170166016, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.0844, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.5339850783348083, | |
| "eval_runtime": 2.0258, | |
| "eval_samples_per_second": 42.946, | |
| "eval_steps_per_second": 1.481, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 64.96, | |
| "eval_accuracy": 0.8735632183908046, | |
| "eval_loss": 0.52364581823349, | |
| "eval_runtime": 2.0251, | |
| "eval_samples_per_second": 42.961, | |
| "eval_steps_per_second": 1.481, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 65.6, | |
| "grad_norm": 10.24560832977295, | |
| "learning_rate": 8.101851851851852e-06, | |
| "loss": 0.0724, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 65.92, | |
| "eval_accuracy": 0.8390804597701149, | |
| "eval_loss": 0.6136645674705505, | |
| "eval_runtime": 2.03, | |
| "eval_samples_per_second": 42.858, | |
| "eval_steps_per_second": 1.478, | |
| "step": 412 | |
| }, | |
| { | |
| "epoch": 66.88, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.5824962854385376, | |
| "eval_runtime": 2.0787, | |
| "eval_samples_per_second": 41.854, | |
| "eval_steps_per_second": 1.443, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 67.2, | |
| "grad_norm": 20.803382873535156, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.0867, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.510515034198761, | |
| "eval_runtime": 2.0565, | |
| "eval_samples_per_second": 42.305, | |
| "eval_steps_per_second": 1.459, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 68.8, | |
| "grad_norm": 15.880162239074707, | |
| "learning_rate": 5.787037037037038e-06, | |
| "loss": 0.071, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 68.96, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5272470116615295, | |
| "eval_runtime": 2.0378, | |
| "eval_samples_per_second": 42.693, | |
| "eval_steps_per_second": 1.472, | |
| "step": 431 | |
| }, | |
| { | |
| "epoch": 69.92, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5523571372032166, | |
| "eval_runtime": 2.0569, | |
| "eval_samples_per_second": 42.297, | |
| "eval_steps_per_second": 1.459, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 70.4, | |
| "grad_norm": 14.639904975891113, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.0723, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 70.88, | |
| "eval_accuracy": 0.8390804597701149, | |
| "eval_loss": 0.5507646799087524, | |
| "eval_runtime": 2.1114, | |
| "eval_samples_per_second": 41.205, | |
| "eval_steps_per_second": 1.421, | |
| "step": 443 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "grad_norm": 6.164122104644775, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.0748, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8160919540229885, | |
| "eval_loss": 0.568942129611969, | |
| "eval_runtime": 2.0852, | |
| "eval_samples_per_second": 41.723, | |
| "eval_steps_per_second": 1.439, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 72.96, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.555583119392395, | |
| "eval_runtime": 2.0316, | |
| "eval_samples_per_second": 42.824, | |
| "eval_steps_per_second": 1.477, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 73.6, | |
| "grad_norm": 11.653559684753418, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.0589, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 73.92, | |
| "eval_accuracy": 0.8505747126436781, | |
| "eval_loss": 0.5452274084091187, | |
| "eval_runtime": 2.0938, | |
| "eval_samples_per_second": 41.551, | |
| "eval_steps_per_second": 1.433, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 74.88, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.5475078225135803, | |
| "eval_runtime": 2.0547, | |
| "eval_samples_per_second": 42.342, | |
| "eval_steps_per_second": 1.46, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 75.2, | |
| "grad_norm": 21.146989822387695, | |
| "learning_rate": 1.1574074074074074e-06, | |
| "loss": 0.0719, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.5483731031417847, | |
| "eval_runtime": 2.1022, | |
| "eval_samples_per_second": 41.386, | |
| "eval_steps_per_second": 1.427, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 76.8, | |
| "grad_norm": 12.87066650390625, | |
| "learning_rate": 0.0, | |
| "loss": 0.0801, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 76.8, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.5496163368225098, | |
| "eval_runtime": 2.0924, | |
| "eval_samples_per_second": 41.58, | |
| "eval_steps_per_second": 1.434, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 76.8, | |
| "step": 480, | |
| "total_flos": 1.514063180200919e+18, | |
| "train_loss": 0.449434948215882, | |
| "train_runtime": 1985.4684, | |
| "train_samples_per_second": 31.549, | |
| "train_steps_per_second": 0.242 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 80, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.514063180200919e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |