| { |
| "best_global_step": 765, |
| "best_metric": 0.06231475993990898, |
| "best_model_checkpoint": "saves_stability/ia3/llama-3-8b-instruct/train_copa_1757340256/checkpoint-765", |
| "epoch": 10.0, |
| "eval_steps": 45, |
| "global_step": 900, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.05555555555555555, |
| "grad_norm": 4.65092658996582, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 0.5534, |
| "num_input_tokens_seen": 1632, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 4.095375061035156, |
| "learning_rate": 5e-06, |
| "loss": 0.6235, |
| "num_input_tokens_seen": 3232, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 4.756162166595459, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 0.6425, |
| "num_input_tokens_seen": 4832, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 4.172198295593262, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 0.7387, |
| "num_input_tokens_seen": 6432, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 4.3731584548950195, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.681, |
| "num_input_tokens_seen": 7968, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 4.341258525848389, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 0.518, |
| "num_input_tokens_seen": 9504, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 3.7260754108428955, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.6923, |
| "num_input_tokens_seen": 11104, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 3.8928024768829346, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.8366, |
| "num_input_tokens_seen": 12704, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 3.961286783218384, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.5993, |
| "num_input_tokens_seen": 14240, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.5326165556907654, |
| "eval_runtime": 0.5815, |
| "eval_samples_per_second": 68.789, |
| "eval_steps_per_second": 17.197, |
| "num_input_tokens_seen": 14240, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 4.945498943328857, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 0.6052, |
| "num_input_tokens_seen": 15808, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 4.686919689178467, |
| "learning_rate": 3e-05, |
| "loss": 0.72, |
| "num_input_tokens_seen": 17344, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 4.145110130310059, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.5328, |
| "num_input_tokens_seen": 18912, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 4.903788089752197, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.6527, |
| "num_input_tokens_seen": 20448, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 3.0253090858459473, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.5486, |
| "num_input_tokens_seen": 21984, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 3.5490806102752686, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.6792, |
| "num_input_tokens_seen": 23552, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 4.2691826820373535, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.6004, |
| "num_input_tokens_seen": 25120, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 4.2351861000061035, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.543, |
| "num_input_tokens_seen": 26656, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 5.384280681610107, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.6312, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.3705243766307831, |
| "eval_runtime": 0.588, |
| "eval_samples_per_second": 68.031, |
| "eval_steps_per_second": 17.008, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 2.4689738750457764, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.2798, |
| "num_input_tokens_seen": 29792, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.8983927369117737, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.2171, |
| "num_input_tokens_seen": 31328, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 1.9559953212738037, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 0.1402, |
| "num_input_tokens_seen": 32832, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 1.7855156660079956, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.0848, |
| "num_input_tokens_seen": 34304, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 0.10905371606349945, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.2313, |
| "num_input_tokens_seen": 35840, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 3.0110857486724854, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.2955, |
| "num_input_tokens_seen": 37376, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 0.355205237865448, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.1038, |
| "num_input_tokens_seen": 38944, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 0.14402715861797333, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.0562, |
| "num_input_tokens_seen": 40512, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 0.14173229038715363, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.0635, |
| "num_input_tokens_seen": 42080, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.07113657146692276, |
| "eval_runtime": 0.5944, |
| "eval_samples_per_second": 67.295, |
| "eval_steps_per_second": 16.824, |
| "num_input_tokens_seen": 42080, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 1.4228240251541138, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.206, |
| "num_input_tokens_seen": 43680, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 0.025234030559659004, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.0488, |
| "num_input_tokens_seen": 45216, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.9494688510894775, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.2344, |
| "num_input_tokens_seen": 46720, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 0.06615747511386871, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.0072, |
| "num_input_tokens_seen": 48288, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 1.4821046590805054, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.1273, |
| "num_input_tokens_seen": 49888, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 0.028987079858779907, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.0373, |
| "num_input_tokens_seen": 51424, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 2.253051280975342, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.215, |
| "num_input_tokens_seen": 52992, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 0.20534025132656097, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.0786, |
| "num_input_tokens_seen": 54592, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.1921127587556839, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.1393, |
| "num_input_tokens_seen": 56192, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.06335774809122086, |
| "eval_runtime": 0.6053, |
| "eval_samples_per_second": 66.088, |
| "eval_steps_per_second": 16.522, |
| "num_input_tokens_seen": 56192, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 0.8517107963562012, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.029, |
| "num_input_tokens_seen": 57760, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 0.020517276600003242, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.0486, |
| "num_input_tokens_seen": 59264, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 0.024442611262202263, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.1048, |
| "num_input_tokens_seen": 60768, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 2.236732006072998, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.1423, |
| "num_input_tokens_seen": 62272, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 0.276692271232605, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.0698, |
| "num_input_tokens_seen": 63808, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 1.8469748497009277, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.205, |
| "num_input_tokens_seen": 65344, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 0.4381294548511505, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.0134, |
| "num_input_tokens_seen": 66912, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 1.713340401649475, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.225, |
| "num_input_tokens_seen": 68480, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 0.3397471010684967, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.0085, |
| "num_input_tokens_seen": 70048, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.06334918737411499, |
| "eval_runtime": 0.5912, |
| "eval_samples_per_second": 67.661, |
| "eval_steps_per_second": 16.915, |
| "num_input_tokens_seen": 70048, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 0.26913413405418396, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0266, |
| "num_input_tokens_seen": 71680, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 2.1941113471984863, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.2459, |
| "num_input_tokens_seen": 73280, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.027516711503267288, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.0303, |
| "num_input_tokens_seen": 74752, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 0.034284986555576324, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.117, |
| "num_input_tokens_seen": 76288, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 0.02439207024872303, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0813, |
| "num_input_tokens_seen": 77888, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 0.15530627965927124, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0067, |
| "num_input_tokens_seen": 79456, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.40555402636528015, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.1203, |
| "num_input_tokens_seen": 81088, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 0.500156581401825, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.0404, |
| "num_input_tokens_seen": 82592, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.6165776252746582, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.1219, |
| "num_input_tokens_seen": 84192, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.06559233367443085, |
| "eval_runtime": 0.5846, |
| "eval_samples_per_second": 68.424, |
| "eval_steps_per_second": 17.106, |
| "num_input_tokens_seen": 84192, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 0.028294576331973076, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.0674, |
| "num_input_tokens_seen": 85696, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.7903892397880554, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.0503, |
| "num_input_tokens_seen": 87296, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 0.11567126214504242, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.031, |
| "num_input_tokens_seen": 88832, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.022067615762352943, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.0224, |
| "num_input_tokens_seen": 90368, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 0.19756221771240234, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.0206, |
| "num_input_tokens_seen": 91968, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.907763659954071, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.0332, |
| "num_input_tokens_seen": 93568, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 0.01762208715081215, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.0359, |
| "num_input_tokens_seen": 95168, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 0.05487053841352463, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.0924, |
| "num_input_tokens_seen": 96704, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 0.023057740181684494, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.1509, |
| "num_input_tokens_seen": 98304, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.06538601219654083, |
| "eval_runtime": 0.5913, |
| "eval_samples_per_second": 67.652, |
| "eval_steps_per_second": 16.913, |
| "num_input_tokens_seen": 98304, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.02764645405113697, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.0052, |
| "num_input_tokens_seen": 99840, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 0.24787607789039612, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.0489, |
| "num_input_tokens_seen": 101376, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 0.42886850237846375, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.1606, |
| "num_input_tokens_seen": 102976, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 1.0214025974273682, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.0785, |
| "num_input_tokens_seen": 104576, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 0.694561243057251, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.2216, |
| "num_input_tokens_seen": 106112, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.024573685601353645, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.1065, |
| "num_input_tokens_seen": 107712, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 0.1497998982667923, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.1273, |
| "num_input_tokens_seen": 109344, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 2.3206191062927246, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.171, |
| "num_input_tokens_seen": 110944, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.5340818762779236, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0373, |
| "num_input_tokens_seen": 112544, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.06596797704696655, |
| "eval_runtime": 0.5849, |
| "eval_samples_per_second": 68.391, |
| "eval_steps_per_second": 17.098, |
| "num_input_tokens_seen": 112544, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.09881038218736649, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.0734, |
| "num_input_tokens_seen": 114144, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 0.8652796745300293, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.0558, |
| "num_input_tokens_seen": 115712, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.6493352055549622, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.1995, |
| "num_input_tokens_seen": 117312, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.8517088890075684, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.1292, |
| "num_input_tokens_seen": 118848, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 1.4027094841003418, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.0525, |
| "num_input_tokens_seen": 120416, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 0.5199717283248901, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0867, |
| "num_input_tokens_seen": 122016, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 1.6734776496887207, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.1298, |
| "num_input_tokens_seen": 123584, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.025618145242333412, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.0165, |
| "num_input_tokens_seen": 125216, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 0.6531297564506531, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.0535, |
| "num_input_tokens_seen": 126784, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.06679404526948929, |
| "eval_runtime": 0.5821, |
| "eval_samples_per_second": 68.711, |
| "eval_steps_per_second": 17.178, |
| "num_input_tokens_seen": 126784, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 0.10721828043460846, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.0243, |
| "num_input_tokens_seen": 128352, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.2870239317417145, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0207, |
| "num_input_tokens_seen": 129920, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 0.08389211446046829, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.0258, |
| "num_input_tokens_seen": 131488, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 0.04516163468360901, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.1243, |
| "num_input_tokens_seen": 133056, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.2532826066017151, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.0535, |
| "num_input_tokens_seen": 134624, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 0.2482670396566391, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0794, |
| "num_input_tokens_seen": 136192, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 0.022973129525780678, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.0384, |
| "num_input_tokens_seen": 137824, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 0.2902039587497711, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.0602, |
| "num_input_tokens_seen": 139392, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.6368235349655151, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.1165, |
| "num_input_tokens_seen": 140960, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.0628441795706749, |
| "eval_runtime": 0.5862, |
| "eval_samples_per_second": 68.235, |
| "eval_steps_per_second": 17.059, |
| "num_input_tokens_seen": 140960, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 0.039155445992946625, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.0875, |
| "num_input_tokens_seen": 142560, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.06545130908489227, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.0679, |
| "num_input_tokens_seen": 144096, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 0.14941030740737915, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.0189, |
| "num_input_tokens_seen": 145632, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 1.0151395797729492, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.0837, |
| "num_input_tokens_seen": 147232, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 0.14988815784454346, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.1201, |
| "num_input_tokens_seen": 148864, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 0.12375698238611221, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.0328, |
| "num_input_tokens_seen": 150432, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 0.5360087156295776, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.0952, |
| "num_input_tokens_seen": 152032, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 0.9888560771942139, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.0481, |
| "num_input_tokens_seen": 153600, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 1.169221043586731, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0766, |
| "num_input_tokens_seen": 155200, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.06539346277713776, |
| "eval_runtime": 0.586, |
| "eval_samples_per_second": 68.259, |
| "eval_steps_per_second": 17.065, |
| "num_input_tokens_seen": 155200, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.037170540541410446, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.1513, |
| "num_input_tokens_seen": 156800, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 0.022741714492440224, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.0093, |
| "num_input_tokens_seen": 158336, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 0.11427551507949829, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0166, |
| "num_input_tokens_seen": 159904, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 1.6933696269989014, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.1117, |
| "num_input_tokens_seen": 161504, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 0.03797967731952667, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.0156, |
| "num_input_tokens_seen": 163008, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 0.45818835496902466, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0718, |
| "num_input_tokens_seen": 164512, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 0.7226192951202393, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.1904, |
| "num_input_tokens_seen": 166080, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.025265689939260483, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.0573, |
| "num_input_tokens_seen": 167680, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.014629139564931393, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.0151, |
| "num_input_tokens_seen": 169216, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.06518089026212692, |
| "eval_runtime": 0.5917, |
| "eval_samples_per_second": 67.606, |
| "eval_steps_per_second": 16.901, |
| "num_input_tokens_seen": 169216, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.04566573724150658, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.0246, |
| "num_input_tokens_seen": 170784, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 0.9680383801460266, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0695, |
| "num_input_tokens_seen": 172352, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.4186362624168396, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.0483, |
| "num_input_tokens_seen": 173920, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.3973574638366699, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0374, |
| "num_input_tokens_seen": 175488, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 1.8178101778030396, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.0804, |
| "num_input_tokens_seen": 177024, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.4705853760242462, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.0542, |
| "num_input_tokens_seen": 178624, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.270990252494812, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.2024, |
| "num_input_tokens_seen": 180096, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 0.9060098528862, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.0598, |
| "num_input_tokens_seen": 181600, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 1.2925699949264526, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.0561, |
| "num_input_tokens_seen": 183232, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.06262358278036118, |
| "eval_runtime": 0.5869, |
| "eval_samples_per_second": 68.159, |
| "eval_steps_per_second": 17.04, |
| "num_input_tokens_seen": 183232, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 0.02199205569922924, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0986, |
| "num_input_tokens_seen": 184768, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 0.01557247806340456, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.083, |
| "num_input_tokens_seen": 186208, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.016952041536569595, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.0291, |
| "num_input_tokens_seen": 187776, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 1.230065941810608, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.0493, |
| "num_input_tokens_seen": 189344, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 0.06948976963758469, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.0219, |
| "num_input_tokens_seen": 190912, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 0.04200341925024986, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.1179, |
| "num_input_tokens_seen": 192512, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.052461329847574234, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.0227, |
| "num_input_tokens_seen": 194112, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.7452025413513184, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.1552, |
| "num_input_tokens_seen": 195712, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.15189005434513092, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0085, |
| "num_input_tokens_seen": 197248, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.06265252828598022, |
| "eval_runtime": 0.5835, |
| "eval_samples_per_second": 68.55, |
| "eval_steps_per_second": 17.137, |
| "num_input_tokens_seen": 197248, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 0.16963563859462738, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.0202, |
| "num_input_tokens_seen": 198784, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 1.5851975679397583, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.1001, |
| "num_input_tokens_seen": 200352, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 0.4539814889431, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.0575, |
| "num_input_tokens_seen": 201856, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 0.02313094586133957, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0757, |
| "num_input_tokens_seen": 203488, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 0.041733697056770325, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.0272, |
| "num_input_tokens_seen": 205056, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 0.27959319949150085, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0891, |
| "num_input_tokens_seen": 206656, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.3239608705043793, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0497, |
| "num_input_tokens_seen": 208288, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 1.5934678316116333, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.1077, |
| "num_input_tokens_seen": 209856, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 1.3411648273468018, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.0554, |
| "num_input_tokens_seen": 211424, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.06260637938976288, |
| "eval_runtime": 0.5897, |
| "eval_samples_per_second": 67.836, |
| "eval_steps_per_second": 16.959, |
| "num_input_tokens_seen": 211424, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 0.0782732367515564, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.0126, |
| "num_input_tokens_seen": 212928, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.13918708264827728, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0308, |
| "num_input_tokens_seen": 214496, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.04588435962796211, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0145, |
| "num_input_tokens_seen": 216064, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 0.033334728330373764, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.1365, |
| "num_input_tokens_seen": 217600, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 1.3441948890686035, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.0571, |
| "num_input_tokens_seen": 219136, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 0.15062227845191956, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0579, |
| "num_input_tokens_seen": 220704, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 0.8167503476142883, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.0962, |
| "num_input_tokens_seen": 222272, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.20099757611751556, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.0178, |
| "num_input_tokens_seen": 223872, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.8290889859199524, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.1764, |
| "num_input_tokens_seen": 225440, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.06353659182786942, |
| "eval_runtime": 0.5843, |
| "eval_samples_per_second": 68.452, |
| "eval_steps_per_second": 17.113, |
| "num_input_tokens_seen": 225440, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 1.780423641204834, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.0747, |
| "num_input_tokens_seen": 226912, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 0.4355238080024719, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.0257, |
| "num_input_tokens_seen": 228512, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.6396685838699341, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.1641, |
| "num_input_tokens_seen": 230080, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.3469223976135254, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.0182, |
| "num_input_tokens_seen": 231616, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 0.01604710891842842, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.0707, |
| "num_input_tokens_seen": 233184, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.7448098063468933, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.0485, |
| "num_input_tokens_seen": 234688, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.2787880301475525, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.1756, |
| "num_input_tokens_seen": 236256, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.19155777990818024, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0337, |
| "num_input_tokens_seen": 237856, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 0.19610635936260223, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.0044, |
| "num_input_tokens_seen": 239392, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.06231475993990898, |
| "eval_runtime": 0.6023, |
| "eval_samples_per_second": 66.417, |
| "eval_steps_per_second": 16.604, |
| "num_input_tokens_seen": 239392, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 0.1609000861644745, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.1528, |
| "num_input_tokens_seen": 240960, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 0.07300806790590286, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.0663, |
| "num_input_tokens_seen": 242496, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 0.7986249923706055, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.0509, |
| "num_input_tokens_seen": 244064, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.0255554411560297, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.0181, |
| "num_input_tokens_seen": 245664, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 0.34656837582588196, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.028, |
| "num_input_tokens_seen": 247264, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 0.33923494815826416, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.0779, |
| "num_input_tokens_seen": 248832, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.0373774990439415, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.1104, |
| "num_input_tokens_seen": 250432, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.04233115166425705, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.0294, |
| "num_input_tokens_seen": 252064, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.01762940175831318, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0226, |
| "num_input_tokens_seen": 253632, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.06255098432302475, |
| "eval_runtime": 0.5768, |
| "eval_samples_per_second": 69.349, |
| "eval_steps_per_second": 17.337, |
| "num_input_tokens_seen": 253632, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.3800027370452881, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.1614, |
| "num_input_tokens_seen": 255232, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 0.037333808839321136, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.0153, |
| "num_input_tokens_seen": 256800, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.014122224412858486, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.1247, |
| "num_input_tokens_seen": 258368, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.0492924265563488, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0118, |
| "num_input_tokens_seen": 259872, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 0.1418236941099167, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.0217, |
| "num_input_tokens_seen": 261376, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 0.6877673268318176, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.048, |
| "num_input_tokens_seen": 262944, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 0.12516207993030548, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.0996, |
| "num_input_tokens_seen": 264544, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.5373877286911011, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.0256, |
| "num_input_tokens_seen": 266080, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 0.46236976981163025, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.0474, |
| "num_input_tokens_seen": 267680, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.06332570314407349, |
| "eval_runtime": 0.5865, |
| "eval_samples_per_second": 68.207, |
| "eval_steps_per_second": 17.052, |
| "num_input_tokens_seen": 267680, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 1.423266053199768, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.2054, |
| "num_input_tokens_seen": 269280, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.7316886782646179, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0369, |
| "num_input_tokens_seen": 270912, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 0.09389762580394745, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.0471, |
| "num_input_tokens_seen": 272512, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 1.3633736371994019, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0516, |
| "num_input_tokens_seen": 274048, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 1.5404667854309082, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.0752, |
| "num_input_tokens_seen": 275648, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 0.03920723497867584, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0788, |
| "num_input_tokens_seen": 277248, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 0.203238844871521, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.047, |
| "num_input_tokens_seen": 278816, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 0.07166783511638641, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.0253, |
| "num_input_tokens_seen": 280416, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.07987658679485321, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.036, |
| "num_input_tokens_seen": 281984, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.06274668872356415, |
| "eval_runtime": 0.5858, |
| "eval_samples_per_second": 68.286, |
| "eval_steps_per_second": 17.071, |
| "num_input_tokens_seen": 281984, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281984, |
| "step": 900, |
| "total_flos": 1.269794471411712e+16, |
| "train_loss": 0.1355803067009482, |
| "train_runtime": 162.3157, |
| "train_samples_per_second": 22.179, |
| "train_steps_per_second": 5.545 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 900, |
| "num_input_tokens_seen": 281984, |
| "num_train_epochs": 10, |
| "save_steps": 45, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.269794471411712e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
