| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.444444444444445, | |
| "eval_steps": 10, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 6.322239398956299, | |
| "learning_rate": 4e-05, | |
| "loss": 2.1156, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "eval_loss": 1.5894473791122437, | |
| "eval_runtime": 9.6219, | |
| "eval_samples_per_second": 10.393, | |
| "eval_steps_per_second": 10.393, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 4.342599391937256, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1893, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "eval_loss": 0.6867849826812744, | |
| "eval_runtime": 9.506, | |
| "eval_samples_per_second": 10.52, | |
| "eval_steps_per_second": 10.52, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 2.0583019256591797, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5218, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "eval_loss": 0.45551854372024536, | |
| "eval_runtime": 9.5056, | |
| "eval_samples_per_second": 10.52, | |
| "eval_steps_per_second": 10.52, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 1.83012056350708, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5292, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "eval_loss": 0.3795148730278015, | |
| "eval_runtime": 9.3637, | |
| "eval_samples_per_second": 10.68, | |
| "eval_steps_per_second": 10.68, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 2.1445064544677734, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3866, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "eval_loss": 0.30648669600486755, | |
| "eval_runtime": 9.2398, | |
| "eval_samples_per_second": 10.823, | |
| "eval_steps_per_second": 10.823, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 1.3674068450927734, | |
| "learning_rate": 0.00019975640502598244, | |
| "loss": 0.3232, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "eval_loss": 0.20737296342849731, | |
| "eval_runtime": 9.2548, | |
| "eval_samples_per_second": 10.805, | |
| "eval_steps_per_second": 10.805, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "grad_norm": 1.0751285552978516, | |
| "learning_rate": 0.00019902680687415705, | |
| "loss": 0.1802, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "eval_loss": 0.15315091609954834, | |
| "eval_runtime": 9.2724, | |
| "eval_samples_per_second": 10.785, | |
| "eval_steps_per_second": 10.785, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.568341076374054, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.21, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "eval_loss": 0.13480396568775177, | |
| "eval_runtime": 9.1987, | |
| "eval_samples_per_second": 10.871, | |
| "eval_steps_per_second": 10.871, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.9985604286193848, | |
| "learning_rate": 0.0001961261695938319, | |
| "loss": 0.158, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.13721750676631927, | |
| "eval_runtime": 9.2578, | |
| "eval_samples_per_second": 10.802, | |
| "eval_steps_per_second": 10.802, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.8096336722373962, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.1629, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "eval_loss": 0.12762245535850525, | |
| "eval_runtime": 9.2566, | |
| "eval_samples_per_second": 10.803, | |
| "eval_steps_per_second": 10.803, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 1.2173279523849487, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.0966, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "eval_loss": 0.10031093657016754, | |
| "eval_runtime": 9.365, | |
| "eval_samples_per_second": 10.678, | |
| "eval_steps_per_second": 10.678, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.7404722571372986, | |
| "learning_rate": 0.00018829475928589271, | |
| "loss": 0.0643, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "eval_loss": 0.08794313669204712, | |
| "eval_runtime": 9.3908, | |
| "eval_samples_per_second": 10.649, | |
| "eval_steps_per_second": 10.649, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1555555555555554, | |
| "grad_norm": 0.9548615217208862, | |
| "learning_rate": 0.0001848048096156426, | |
| "loss": 0.0726, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.1555555555555554, | |
| "eval_loss": 0.08720792084932327, | |
| "eval_runtime": 9.3262, | |
| "eval_samples_per_second": 10.722, | |
| "eval_steps_per_second": 10.722, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.2444444444444445, | |
| "grad_norm": 1.0252933502197266, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0493, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2444444444444445, | |
| "eval_loss": 0.09057007730007172, | |
| "eval_runtime": 9.3221, | |
| "eval_samples_per_second": 10.727, | |
| "eval_steps_per_second": 10.727, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.5175765156745911, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 0.0746, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "eval_loss": 0.058685798197984695, | |
| "eval_runtime": 9.3758, | |
| "eval_samples_per_second": 10.666, | |
| "eval_steps_per_second": 10.666, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.33382534980773926, | |
| "learning_rate": 0.0001719339800338651, | |
| "loss": 0.0473, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "eval_loss": 0.0560651533305645, | |
| "eval_runtime": 9.6017, | |
| "eval_samples_per_second": 10.415, | |
| "eval_steps_per_second": 10.415, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.511111111111111, | |
| "grad_norm": 1.0913978815078735, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.0644, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.511111111111111, | |
| "eval_loss": 0.05025744438171387, | |
| "eval_runtime": 9.3986, | |
| "eval_samples_per_second": 10.64, | |
| "eval_steps_per_second": 10.64, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.8013315796852112, | |
| "learning_rate": 0.0001615661475325658, | |
| "loss": 0.0366, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.030684156343340874, | |
| "eval_runtime": 9.3927, | |
| "eval_samples_per_second": 10.647, | |
| "eval_steps_per_second": 10.647, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6888888888888889, | |
| "grad_norm": 0.2898242175579071, | |
| "learning_rate": 0.0001559192903470747, | |
| "loss": 0.0247, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.6888888888888889, | |
| "eval_loss": 0.023328043520450592, | |
| "eval_runtime": 9.6997, | |
| "eval_samples_per_second": 10.31, | |
| "eval_steps_per_second": 10.31, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.0667782798409462, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.01, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "eval_loss": 0.021545417606830597, | |
| "eval_runtime": 9.6129, | |
| "eval_samples_per_second": 10.403, | |
| "eval_steps_per_second": 10.403, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8666666666666667, | |
| "grad_norm": 1.8411855697631836, | |
| "learning_rate": 0.00014383711467890774, | |
| "loss": 0.0393, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.8666666666666667, | |
| "eval_loss": 0.012232878245413303, | |
| "eval_runtime": 9.4898, | |
| "eval_samples_per_second": 10.538, | |
| "eval_steps_per_second": 10.538, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.9555555555555557, | |
| "grad_norm": 2.9970459938049316, | |
| "learning_rate": 0.00013746065934159123, | |
| "loss": 0.0299, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.9555555555555557, | |
| "eval_loss": 0.01798514649271965, | |
| "eval_runtime": 9.5905, | |
| "eval_samples_per_second": 10.427, | |
| "eval_steps_per_second": 10.427, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0444444444444443, | |
| "grad_norm": 1.0432640314102173, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0166, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0444444444444443, | |
| "eval_loss": 0.008207106962800026, | |
| "eval_runtime": 9.2773, | |
| "eval_samples_per_second": 10.779, | |
| "eval_steps_per_second": 10.779, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 1.9301716089248657, | |
| "learning_rate": 0.00012419218955996676, | |
| "loss": 0.0319, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "eval_loss": 0.008276881650090218, | |
| "eval_runtime": 9.327, | |
| "eval_samples_per_second": 10.722, | |
| "eval_steps_per_second": 10.722, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.07122544944286346, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.0077, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "eval_loss": 0.007150276098400354, | |
| "eval_runtime": 9.3785, | |
| "eval_samples_per_second": 10.663, | |
| "eval_steps_per_second": 10.663, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.311111111111111, | |
| "grad_norm": 0.16882538795471191, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.0141, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.311111111111111, | |
| "eval_loss": 0.003109171986579895, | |
| "eval_runtime": 9.3703, | |
| "eval_samples_per_second": 10.672, | |
| "eval_steps_per_second": 10.672, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.037282537668943405, | |
| "learning_rate": 0.00010348994967025012, | |
| "loss": 0.0017, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.012033730745315552, | |
| "eval_runtime": 9.2982, | |
| "eval_samples_per_second": 10.755, | |
| "eval_steps_per_second": 10.755, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.022591086104512215, | |
| "learning_rate": 9.651005032974994e-05, | |
| "loss": 0.0015, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "eval_loss": 0.015280201099812984, | |
| "eval_runtime": 9.4013, | |
| "eval_samples_per_second": 10.637, | |
| "eval_steps_per_second": 10.637, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.5777777777777775, | |
| "grad_norm": 1.2939496040344238, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.0126, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.5777777777777775, | |
| "eval_loss": 0.01406156551092863, | |
| "eval_runtime": 9.3181, | |
| "eval_samples_per_second": 10.732, | |
| "eval_steps_per_second": 10.732, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.009937470778822899, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.0043, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "eval_loss": 0.0021963752806186676, | |
| "eval_runtime": 9.3813, | |
| "eval_samples_per_second": 10.659, | |
| "eval_steps_per_second": 10.659, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7555555555555555, | |
| "grad_norm": 0.022694729268550873, | |
| "learning_rate": 7.580781044003324e-05, | |
| "loss": 0.0068, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.7555555555555555, | |
| "eval_loss": 0.001871286309324205, | |
| "eval_runtime": 9.7104, | |
| "eval_samples_per_second": 10.298, | |
| "eval_steps_per_second": 10.298, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.11103329062461853, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0018, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "eval_loss": 0.002184124430641532, | |
| "eval_runtime": 9.5703, | |
| "eval_samples_per_second": 10.449, | |
| "eval_steps_per_second": 10.449, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.9333333333333336, | |
| "grad_norm": 0.02097630314528942, | |
| "learning_rate": 6.25393406584088e-05, | |
| "loss": 0.0026, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.9333333333333336, | |
| "eval_loss": 0.0034216546919196844, | |
| "eval_runtime": 9.647, | |
| "eval_samples_per_second": 10.366, | |
| "eval_steps_per_second": 10.366, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.022222222222222, | |
| "grad_norm": 0.0426625981926918, | |
| "learning_rate": 5.616288532109225e-05, | |
| "loss": 0.0017, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.022222222222222, | |
| "eval_loss": 0.007565508596599102, | |
| "eval_runtime": 9.5797, | |
| "eval_samples_per_second": 10.439, | |
| "eval_steps_per_second": 10.439, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.111111111111111, | |
| "grad_norm": 0.0036587081849575043, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.0002, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.111111111111111, | |
| "eval_loss": 0.010203778743743896, | |
| "eval_runtime": 9.5558, | |
| "eval_samples_per_second": 10.465, | |
| "eval_steps_per_second": 10.465, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.008132525719702244, | |
| "learning_rate": 4.4080709652925336e-05, | |
| "loss": 0.0004, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.011154056526720524, | |
| "eval_runtime": 9.3832, | |
| "eval_samples_per_second": 10.657, | |
| "eval_steps_per_second": 10.657, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.2888888888888888, | |
| "grad_norm": 0.00727389520034194, | |
| "learning_rate": 3.843385246743417e-05, | |
| "loss": 0.006, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.2888888888888888, | |
| "eval_loss": 0.00937813799828291, | |
| "eval_runtime": 9.3745, | |
| "eval_samples_per_second": 10.667, | |
| "eval_steps_per_second": 10.667, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.3777777777777778, | |
| "grad_norm": 0.004803340416401625, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.0003, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.3777777777777778, | |
| "eval_loss": 0.007455301936715841, | |
| "eval_runtime": 9.5178, | |
| "eval_samples_per_second": 10.507, | |
| "eval_steps_per_second": 10.507, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.466666666666667, | |
| "grad_norm": 0.005293034482747316, | |
| "learning_rate": 2.8066019966134904e-05, | |
| "loss": 0.0003, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.466666666666667, | |
| "eval_loss": 0.006887929514050484, | |
| "eval_runtime": 9.5617, | |
| "eval_samples_per_second": 10.458, | |
| "eval_steps_per_second": 10.458, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.01216947752982378, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.0002, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.006745634134858847, | |
| "eval_runtime": 9.5509, | |
| "eval_samples_per_second": 10.47, | |
| "eval_steps_per_second": 10.47, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.6444444444444444, | |
| "grad_norm": 0.008572138845920563, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0005, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.6444444444444444, | |
| "eval_loss": 0.006609635427594185, | |
| "eval_runtime": 9.6817, | |
| "eval_samples_per_second": 10.329, | |
| "eval_steps_per_second": 10.329, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.7333333333333334, | |
| "grad_norm": 0.004443774465471506, | |
| "learning_rate": 1.5195190384357404e-05, | |
| "loss": 0.0003, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.7333333333333334, | |
| "eval_loss": 0.007157918065786362, | |
| "eval_runtime": 9.5559, | |
| "eval_samples_per_second": 10.465, | |
| "eval_steps_per_second": 10.465, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.822222222222222, | |
| "grad_norm": 0.0011405730620026588, | |
| "learning_rate": 1.1705240714107302e-05, | |
| "loss": 0.0037, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.822222222222222, | |
| "eval_loss": 0.006302958354353905, | |
| "eval_runtime": 9.2431, | |
| "eval_samples_per_second": 10.819, | |
| "eval_steps_per_second": 10.819, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.004764024633914232, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.004, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "eval_loss": 0.005341523326933384, | |
| "eval_runtime": 9.5532, | |
| "eval_samples_per_second": 10.468, | |
| "eval_steps_per_second": 10.468, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.004020992666482925, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 0.0003, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.005209792871028185, | |
| "eval_runtime": 9.7502, | |
| "eval_samples_per_second": 10.256, | |
| "eval_steps_per_second": 10.256, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.088888888888889, | |
| "grad_norm": 0.013617518357932568, | |
| "learning_rate": 3.873830406168111e-06, | |
| "loss": 0.0002, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.088888888888889, | |
| "eval_loss": 0.005074501037597656, | |
| "eval_runtime": 9.3359, | |
| "eval_samples_per_second": 10.711, | |
| "eval_steps_per_second": 10.711, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.177777777777778, | |
| "grad_norm": 0.003857893170788884, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.0002, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.177777777777778, | |
| "eval_loss": 0.004982742480933666, | |
| "eval_runtime": 9.5873, | |
| "eval_samples_per_second": 10.43, | |
| "eval_steps_per_second": 10.43, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.006583555601537228, | |
| "learning_rate": 9.731931258429638e-07, | |
| "loss": 0.0006, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "eval_loss": 0.004874282516539097, | |
| "eval_runtime": 9.6607, | |
| "eval_samples_per_second": 10.351, | |
| "eval_steps_per_second": 10.351, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.355555555555555, | |
| "grad_norm": 0.004742850083857775, | |
| "learning_rate": 2.4359497401758024e-07, | |
| "loss": 0.0005, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.355555555555555, | |
| "eval_loss": 0.004786411300301552, | |
| "eval_runtime": 9.6649, | |
| "eval_samples_per_second": 10.347, | |
| "eval_steps_per_second": 10.347, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 0.010840805247426033, | |
| "learning_rate": 0.0, | |
| "loss": 0.0002, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "eval_loss": 0.005105508491396904, | |
| "eval_runtime": 9.3806, | |
| "eval_samples_per_second": 10.66, | |
| "eval_steps_per_second": 10.66, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "step": 500, | |
| "total_flos": 1.038706354200576e+16, | |
| "train_loss": 0.1301481350355316, | |
| "train_runtime": 1904.5573, | |
| "train_samples_per_second": 2.1, | |
| "train_steps_per_second": 0.263 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.038706354200576e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |