| { |
| "best_global_step": 270, |
| "best_metric": 0.05592306703329086, |
| "best_model_checkpoint": "saves/p-tuning/llama-3-8b-instruct/train_copa_1754648732/checkpoint-270", |
| "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": 182.23764038085938, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 3.024, |
| "num_input_tokens_seen": 1536, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 27.16082763671875, |
| "learning_rate": 5e-06, |
| "loss": 0.595, |
| "num_input_tokens_seen": 3168, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 0.7717798948287964, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 0.1449, |
| "num_input_tokens_seen": 4736, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 4.796186447143555, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 0.1717, |
| "num_input_tokens_seen": 6304, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 267.90338134765625, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.7707, |
| "num_input_tokens_seen": 7840, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 79.07738494873047, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 4.1646, |
| "num_input_tokens_seen": 9408, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 160.53985595703125, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.994, |
| "num_input_tokens_seen": 10912, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 9.1184720993042, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.4008, |
| "num_input_tokens_seen": 12448, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 4.6666388511657715, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.2698, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.2506703734397888, |
| "eval_runtime": 1.5555, |
| "eval_samples_per_second": 25.716, |
| "eval_steps_per_second": 6.429, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 13.625567436218262, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 1.0066, |
| "num_input_tokens_seen": 15584, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 66.29561614990234, |
| "learning_rate": 3e-05, |
| "loss": 0.4067, |
| "num_input_tokens_seen": 17184, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 4.158051490783691, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.2742, |
| "num_input_tokens_seen": 18752, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 8.074698448181152, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.2676, |
| "num_input_tokens_seen": 20352, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 1.2780441045761108, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.2568, |
| "num_input_tokens_seen": 21952, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 3.60463809967041, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.2286, |
| "num_input_tokens_seen": 23456, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 1.2155777215957642, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.2512, |
| "num_input_tokens_seen": 25056, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 1.3992754220962524, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.2411, |
| "num_input_tokens_seen": 26560, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.5974087715148926, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.2395, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.3042182922363281, |
| "eval_runtime": 1.5554, |
| "eval_samples_per_second": 25.717, |
| "eval_steps_per_second": 6.429, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 0.2746608853340149, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.2761, |
| "num_input_tokens_seen": 29696, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.22459860146045685, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.2241, |
| "num_input_tokens_seen": 31232, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 203.96246337890625, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 1.2972, |
| "num_input_tokens_seen": 32768, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 6.3060150146484375, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.4149, |
| "num_input_tokens_seen": 34304, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 0.7371984124183655, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.0178, |
| "num_input_tokens_seen": 35872, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 0.9288480877876282, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.3059, |
| "num_input_tokens_seen": 37408, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 5.491672992706299, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.1848, |
| "num_input_tokens_seen": 38976, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 1.3010174036026, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.0486, |
| "num_input_tokens_seen": 40576, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 1.7566763162612915, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.1339, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.16016362607479095, |
| "eval_runtime": 1.6221, |
| "eval_samples_per_second": 24.66, |
| "eval_steps_per_second": 6.165, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 0.7288216948509216, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.0618, |
| "num_input_tokens_seen": 43680, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 1.4960017204284668, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.2294, |
| "num_input_tokens_seen": 45248, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.010506484657526016, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.0312, |
| "num_input_tokens_seen": 46816, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 4.493790149688721, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.2127, |
| "num_input_tokens_seen": 48384, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 0.052334196865558624, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.014, |
| "num_input_tokens_seen": 49952, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 0.007493265438824892, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.0074, |
| "num_input_tokens_seen": 51520, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 0.11316759139299393, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.0089, |
| "num_input_tokens_seen": 53024, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 3.800060987472534, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.1824, |
| "num_input_tokens_seen": 54592, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 6.699803352355957, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.0821, |
| "num_input_tokens_seen": 56128, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.07743295282125473, |
| "eval_runtime": 1.5826, |
| "eval_samples_per_second": 25.276, |
| "eval_steps_per_second": 6.319, |
| "num_input_tokens_seen": 56128, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 0.02711506374180317, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.0294, |
| "num_input_tokens_seen": 57696, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 5.267457485198975, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.1222, |
| "num_input_tokens_seen": 59264, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 0.128107950091362, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.0042, |
| "num_input_tokens_seen": 60864, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 0.3165455758571625, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.0125, |
| "num_input_tokens_seen": 62464, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 3.6155238151550293, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.0905, |
| "num_input_tokens_seen": 64032, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 0.8466137647628784, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.0133, |
| "num_input_tokens_seen": 65568, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 0.03307263180613518, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.0061, |
| "num_input_tokens_seen": 67104, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 0.004339626524597406, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.0357, |
| "num_input_tokens_seen": 68704, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 4.743404388427734, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.0255, |
| "num_input_tokens_seen": 70272, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.10026385635137558, |
| "eval_runtime": 1.5971, |
| "eval_samples_per_second": 25.046, |
| "eval_steps_per_second": 6.262, |
| "num_input_tokens_seen": 70272, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 0.016594767570495605, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 71808, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 13.802270889282227, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.2865, |
| "num_input_tokens_seen": 73408, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 1.540993094444275, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.0897, |
| "num_input_tokens_seen": 74912, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 3.656810760498047, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.0803, |
| "num_input_tokens_seen": 76544, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 0.20033599436283112, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 78112, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 0.08203268051147461, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0024, |
| "num_input_tokens_seen": 79712, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.020763523876667023, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.0041, |
| "num_input_tokens_seen": 81280, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 0.009530600160360336, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.1335, |
| "num_input_tokens_seen": 82848, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.06433311849832535, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.0231, |
| "num_input_tokens_seen": 84352, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.05592306703329086, |
| "eval_runtime": 1.5767, |
| "eval_samples_per_second": 25.37, |
| "eval_steps_per_second": 6.342, |
| "num_input_tokens_seen": 84352, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 0.020673708990216255, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 85920, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.10938331484794617, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.0012, |
| "num_input_tokens_seen": 87520, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 0.005322723183780909, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 89088, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.09109042584896088, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.1136, |
| "num_input_tokens_seen": 90688, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 0.010418006218969822, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.0137, |
| "num_input_tokens_seen": 92160, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.94423508644104, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.007, |
| "num_input_tokens_seen": 93760, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 9.637407302856445, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.0611, |
| "num_input_tokens_seen": 95360, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 0.0038525161799043417, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.0445, |
| "num_input_tokens_seen": 96928, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 0.07096295803785324, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.0061, |
| "num_input_tokens_seen": 98464, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.11320608854293823, |
| "eval_runtime": 1.6079, |
| "eval_samples_per_second": 24.877, |
| "eval_steps_per_second": 6.219, |
| "num_input_tokens_seen": 98464, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.2531680464744568, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.0421, |
| "num_input_tokens_seen": 100064, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 0.07123122364282608, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.0156, |
| "num_input_tokens_seen": 101600, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 2.079775333404541, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.0438, |
| "num_input_tokens_seen": 103200, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 0.16428519785404205, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.0037, |
| "num_input_tokens_seen": 104768, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 0.006941991858184338, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.096, |
| "num_input_tokens_seen": 106304, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.0025284558068960905, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 107904, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 0.3235416114330292, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 109408, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 0.8172513842582703, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.0038, |
| "num_input_tokens_seen": 111008, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.0009636105969548225, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 112576, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.0910610780119896, |
| "eval_runtime": 1.4599, |
| "eval_samples_per_second": 27.399, |
| "eval_steps_per_second": 6.85, |
| "num_input_tokens_seen": 112576, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.2835381329059601, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.0113, |
| "num_input_tokens_seen": 114144, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 0.2145785391330719, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 115712, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.0012069224612787366, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.022, |
| "num_input_tokens_seen": 117216, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.002359800273552537, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 118816, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 0.15576356649398804, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 120352, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 0.0018543049227446318, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 121920, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 9.92176342010498, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.0205, |
| "num_input_tokens_seen": 123456, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.010661592707037926, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 125056, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 1.2817062139511108, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.0059, |
| "num_input_tokens_seen": 126624, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.0957321971654892, |
| "eval_runtime": 1.5931, |
| "eval_samples_per_second": 25.109, |
| "eval_steps_per_second": 6.277, |
| "num_input_tokens_seen": 126624, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 0.005725174676626921, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.001, |
| "num_input_tokens_seen": 128224, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.004600477870553732, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 129728, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 0.003879501484334469, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 131328, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 0.003114352934062481, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 132896, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.004052264150232077, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 134496, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 0.0023121682461351156, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 136064, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 0.002999385353177786, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 137664, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 0.0015907135093584657, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 139232, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.0016937963664531708, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 140832, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.11632829904556274, |
| "eval_runtime": 1.5605, |
| "eval_samples_per_second": 25.633, |
| "eval_steps_per_second": 6.408, |
| "num_input_tokens_seen": 140832, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 0.0018119423184543848, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 142368, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.0016430102987214923, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 144000, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 0.00153221539221704, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 145632, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 0.0023474155459553003, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 147168, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 0.0013369086664170027, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 148736, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 0.0018505568150430918, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 150304, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 0.003105954499915242, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 151872, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 0.0011468834709376097, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 153472, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 0.0012695087352767587, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 154976, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.11310011148452759, |
| "eval_runtime": 1.5808, |
| "eval_samples_per_second": 25.304, |
| "eval_steps_per_second": 6.326, |
| "num_input_tokens_seen": 154976, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.0010720669524744153, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 156544, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 0.0008880237000994384, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 158112, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 0.010376314632594585, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 159680, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 0.0012880179565399885, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 161312, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 0.0007874983130022883, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 162880, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 0.0006399190169759095, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 164448, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 0.0007795878336764872, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 166016, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.0007273498922586441, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 167552, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.0009823073633015156, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 169056, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.11383338272571564, |
| "eval_runtime": 1.5697, |
| "eval_samples_per_second": 25.483, |
| "eval_steps_per_second": 6.371, |
| "num_input_tokens_seen": 169056, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.0006269391742534935, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 170592, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 0.004947153385728598, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 172192, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.0008867355645634234, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 173792, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.0006600656779482961, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 175360, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 0.0006670368602499366, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 176992, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.0006873439997434616, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 178592, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.0005355999455787241, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 180128, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 0.0034590065479278564, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 181632, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 0.0038449957501143217, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 183200, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.11220748722553253, |
| "eval_runtime": 1.621, |
| "eval_samples_per_second": 24.677, |
| "eval_steps_per_second": 6.169, |
| "num_input_tokens_seen": 183200, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 0.00060513976495713, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 184800, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 0.0006962447660043836, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 186336, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.0006690849550068378, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 187904, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 0.00067218451295048, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 189472, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 0.0006238952628336847, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 191072, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 0.000612985750194639, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 192608, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.0005657134461216629, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 194208, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.0004337099671829492, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 195776, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.000535633647814393, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 197344, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.11814818531274796, |
| "eval_runtime": 1.5634, |
| "eval_samples_per_second": 25.585, |
| "eval_steps_per_second": 6.396, |
| "num_input_tokens_seen": 197344, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 0.0005506157176569104, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 198944, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 0.000527763448189944, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 200512, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 0.00044364368659444153, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 202016, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 0.0010423212079331279, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 203648, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 0.0004312348610255867, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 205184, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 0.0007077080081216991, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 206720, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.0005013030604459345, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 208320, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 0.0006409093039110303, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 209856, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 0.0006091762334108353, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 211392, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.113214410841465, |
| "eval_runtime": 1.5773, |
| "eval_samples_per_second": 25.361, |
| "eval_steps_per_second": 6.34, |
| "num_input_tokens_seen": 211392, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 0.0005878661759197712, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 212960, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.0036648926325142384, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 214528, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.0005150093347765505, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 216000, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 0.00036882495624013245, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 217632, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 0.0006515193963423371, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 219232, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 0.00045884872088208795, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 220800, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 0.0006991951377131045, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 222368, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.009923174045979977, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 223968, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.00046975447912700474, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 225536, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.11708308756351471, |
| "eval_runtime": 1.5603, |
| "eval_samples_per_second": 25.636, |
| "eval_steps_per_second": 6.409, |
| "num_input_tokens_seen": 225536, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 0.0005550209898501635, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 227168, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 0.0006263662362471223, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 228704, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.00046402690350078046, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 230336, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.0008687042281962931, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 231936, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 0.005186337511986494, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 233472, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.0005630331579595804, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 235040, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.00039121092413552105, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 236608, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.00047918211203068495, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 238144, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 0.008829047903418541, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 239680, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.11304056644439697, |
| "eval_runtime": 1.5994, |
| "eval_samples_per_second": 25.009, |
| "eval_steps_per_second": 6.252, |
| "num_input_tokens_seen": 239680, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 0.0005217266152612865, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 241280, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 0.00043171996367163956, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 242816, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 0.0006213925080373883, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 244352, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.0004608691670000553, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 245856, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 0.00039821735117584467, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 247456, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 0.00043877094867639244, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 248992, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.00038490648148581386, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 250528, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.0008438437944278121, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 252128, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.0004914125893265009, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 253696, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.11710908263921738, |
| "eval_runtime": 1.5994, |
| "eval_samples_per_second": 25.009, |
| "eval_steps_per_second": 6.252, |
| "num_input_tokens_seen": 253696, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.0004190673353150487, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 255296, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 0.00036627528606913984, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 256896, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.00042660304461605847, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 258432, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.0014467159053310752, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 260000, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 0.0004888187977485359, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 261568, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 0.0005717790336348116, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 263200, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 0.0010923414956778288, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 264736, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.0005615805857814848, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 266304, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 0.0004197074449621141, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 267840, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.11252784729003906, |
| "eval_runtime": 1.4701, |
| "eval_samples_per_second": 27.209, |
| "eval_steps_per_second": 6.802, |
| "num_input_tokens_seen": 267840, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 0.0004339388688094914, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 269376, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.0005315785529091954, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 270944, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 0.0007396696018986404, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 272448, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.0004039212071802467, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 273984, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 0.00044190537300892174, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 275520, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 0.0004263962327968329, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 277152, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 0.0004176957008894533, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 278688, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 0.0004536750493571162, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.0, |
| "num_input_tokens_seen": 280256, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.0004271083453204483, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.0, |
| "num_input_tokens_seen": 281856, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.11861652135848999, |
| "eval_runtime": 1.5879, |
| "eval_samples_per_second": 25.191, |
| "eval_steps_per_second": 6.298, |
| "num_input_tokens_seen": 281856, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281856, |
| "step": 900, |
| "total_flos": 1.2691848290107392e+16, |
| "train_loss": 0.1051763629641856, |
| "train_runtime": 637.0083, |
| "train_samples_per_second": 5.651, |
| "train_steps_per_second": 1.413 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 900, |
| "num_input_tokens_seen": 281856, |
| "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": 1.2691848290107392e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
