| { | |
| "best_metric": 0.13376566767692566, | |
| "best_model_checkpoint": "/media/user/Expansion/flan-t5-small-capitalizer/checkpoint-50000", | |
| "epoch": 1.0, | |
| "eval_steps": 2500, | |
| "global_step": 50000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 1.4570035934448242, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.3309, | |
| "num_input_tokens_seen": 570534, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.5052685737609863, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.2655, | |
| "num_input_tokens_seen": 1140292, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.3262746334075928, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.2673, | |
| "num_input_tokens_seen": 1697650, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.0782378911972046, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.2606, | |
| "num_input_tokens_seen": 2254526, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.1907426118850708, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.2532, | |
| "num_input_tokens_seen": 2824810, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.1738843023777008, | |
| "eval_runtime": 44.8865, | |
| "eval_samples_per_second": 55.696, | |
| "eval_steps_per_second": 6.973, | |
| "num_input_tokens_seen": 2824810, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 2.4040281772613525, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.2395, | |
| "num_input_tokens_seen": 3398362, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.624157428741455, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.2372, | |
| "num_input_tokens_seen": 3976908, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.1766471862792969, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.2333, | |
| "num_input_tokens_seen": 4534622, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 1.0758516788482666, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.2244, | |
| "num_input_tokens_seen": 5111014, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.210937261581421, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.231, | |
| "num_input_tokens_seen": 5702148, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.16534267365932465, | |
| "eval_runtime": 44.9542, | |
| "eval_samples_per_second": 55.612, | |
| "eval_steps_per_second": 6.963, | |
| "num_input_tokens_seen": 5702148, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.975110411643982, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.2232, | |
| "num_input_tokens_seen": 6276610, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 2.1255569458007812, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.2184, | |
| "num_input_tokens_seen": 6847280, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 1.8061076402664185, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.216, | |
| "num_input_tokens_seen": 7406418, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 1.6190805435180664, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.2346, | |
| "num_input_tokens_seen": 7976560, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.7121838331222534, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.2163, | |
| "num_input_tokens_seen": 8531178, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.15705962479114532, | |
| "eval_runtime": 45.3915, | |
| "eval_samples_per_second": 55.076, | |
| "eval_steps_per_second": 6.896, | |
| "num_input_tokens_seen": 8531178, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.3726420402526855, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.213, | |
| "num_input_tokens_seen": 9096822, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 1.3163851499557495, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.2152, | |
| "num_input_tokens_seen": 9655322, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.5095873475074768, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.2151, | |
| "num_input_tokens_seen": 10222666, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.6344404220581055, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.2097, | |
| "num_input_tokens_seen": 10783398, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.9461726546287537, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1966, | |
| "num_input_tokens_seen": 11350902, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.1528666764497757, | |
| "eval_runtime": 45.1903, | |
| "eval_samples_per_second": 55.322, | |
| "eval_steps_per_second": 6.926, | |
| "num_input_tokens_seen": 11350902, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.5499858260154724, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.202, | |
| "num_input_tokens_seen": 11937536, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.9817250967025757, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.2055, | |
| "num_input_tokens_seen": 12530266, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 1.1686115264892578, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.2045, | |
| "num_input_tokens_seen": 13084430, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.8986425399780273, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.2005, | |
| "num_input_tokens_seen": 13645732, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.1480282545089722, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2013, | |
| "num_input_tokens_seen": 14191502, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.14906854927539825, | |
| "eval_runtime": 47.8025, | |
| "eval_samples_per_second": 52.299, | |
| "eval_steps_per_second": 6.548, | |
| "num_input_tokens_seen": 14191502, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.8384343981742859, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.2027, | |
| "num_input_tokens_seen": 14784072, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 1.3071606159210205, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.2068, | |
| "num_input_tokens_seen": 15336696, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.5159173011779785, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.2093, | |
| "num_input_tokens_seen": 15914526, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 1.51547110080719, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.1908, | |
| "num_input_tokens_seen": 16486110, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.7921791076660156, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.1971, | |
| "num_input_tokens_seen": 17050704, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.14638836681842804, | |
| "eval_runtime": 46.342, | |
| "eval_samples_per_second": 53.947, | |
| "eval_steps_per_second": 6.754, | |
| "num_input_tokens_seen": 17050704, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.5466729402542114, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.1948, | |
| "num_input_tokens_seen": 17650510, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.6244600415229797, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.1863, | |
| "num_input_tokens_seen": 18201474, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 1.0302484035491943, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.185, | |
| "num_input_tokens_seen": 18734382, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.45665842294692993, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.1883, | |
| "num_input_tokens_seen": 19293596, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.8066365122795105, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.1791, | |
| "num_input_tokens_seen": 19857804, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.14467556774616241, | |
| "eval_runtime": 45.168, | |
| "eval_samples_per_second": 55.349, | |
| "eval_steps_per_second": 6.93, | |
| "num_input_tokens_seen": 19857804, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 6.650664329528809, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.1929, | |
| "num_input_tokens_seen": 20418316, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.9265847206115723, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.1864, | |
| "num_input_tokens_seen": 20975738, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.9065707921981812, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.1783, | |
| "num_input_tokens_seen": 21552842, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 1.3064478635787964, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.1849, | |
| "num_input_tokens_seen": 22115314, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.7317535281181335, | |
| "learning_rate": 3e-05, | |
| "loss": 0.193, | |
| "num_input_tokens_seen": 22687180, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.14237256348133087, | |
| "eval_runtime": 46.1002, | |
| "eval_samples_per_second": 54.23, | |
| "eval_steps_per_second": 6.79, | |
| "num_input_tokens_seen": 22687180, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 1.0536916255950928, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.1875, | |
| "num_input_tokens_seen": 23265246, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.7591565847396851, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.1775, | |
| "num_input_tokens_seen": 23837722, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.7889916300773621, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.1833, | |
| "num_input_tokens_seen": 24407840, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 2.9207775592803955, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.1901, | |
| "num_input_tokens_seen": 24987106, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.3993462026119232, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.1821, | |
| "num_input_tokens_seen": 25532518, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.14162951707839966, | |
| "eval_runtime": 45.1335, | |
| "eval_samples_per_second": 55.391, | |
| "eval_steps_per_second": 6.935, | |
| "num_input_tokens_seen": 25532518, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 1.0471628904342651, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.1898, | |
| "num_input_tokens_seen": 26101602, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.4212801158428192, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.177, | |
| "num_input_tokens_seen": 26700486, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.0187805891036987, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.1932, | |
| "num_input_tokens_seen": 27290876, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.48242756724357605, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.1797, | |
| "num_input_tokens_seen": 27851074, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.9889593720436096, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.19, | |
| "num_input_tokens_seen": 28423408, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.13971897959709167, | |
| "eval_runtime": 45.1742, | |
| "eval_samples_per_second": 55.341, | |
| "eval_steps_per_second": 6.929, | |
| "num_input_tokens_seen": 28423408, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.6578853726387024, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.191, | |
| "num_input_tokens_seen": 28986964, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.1678123474121094, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.1807, | |
| "num_input_tokens_seen": 29555682, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.3844729959964752, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.174, | |
| "num_input_tokens_seen": 30106292, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 1.3147269487380981, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.184, | |
| "num_input_tokens_seen": 30669702, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.0336247682571411, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1753, | |
| "num_input_tokens_seen": 31248170, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.13881850242614746, | |
| "eval_runtime": 45.1907, | |
| "eval_samples_per_second": 55.321, | |
| "eval_steps_per_second": 6.926, | |
| "num_input_tokens_seen": 31248170, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.407716989517212, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.1725, | |
| "num_input_tokens_seen": 31809958, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 1.2787240743637085, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.1811, | |
| "num_input_tokens_seen": 32366844, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.9074501395225525, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.184, | |
| "num_input_tokens_seen": 32913758, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.5810595154762268, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.1754, | |
| "num_input_tokens_seen": 33479540, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.109554648399353, | |
| "learning_rate": 2e-05, | |
| "loss": 0.184, | |
| "num_input_tokens_seen": 34048604, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.13782289624214172, | |
| "eval_runtime": 45.1441, | |
| "eval_samples_per_second": 55.378, | |
| "eval_steps_per_second": 6.933, | |
| "num_input_tokens_seen": 34048604, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.1160213947296143, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.1674, | |
| "num_input_tokens_seen": 34606292, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.3033101558685303, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.1817, | |
| "num_input_tokens_seen": 35171758, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.44109177589416504, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.1854, | |
| "num_input_tokens_seen": 35716556, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.6188114285469055, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.1793, | |
| "num_input_tokens_seen": 36298888, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.612743079662323, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1717, | |
| "num_input_tokens_seen": 36903282, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.13706056773662567, | |
| "eval_runtime": 45.1443, | |
| "eval_samples_per_second": 55.378, | |
| "eval_steps_per_second": 6.933, | |
| "num_input_tokens_seen": 36903282, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.1475753784179688, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.1814, | |
| "num_input_tokens_seen": 37471044, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.7694216966629028, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.1813, | |
| "num_input_tokens_seen": 38025842, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.7751884460449219, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.1753, | |
| "num_input_tokens_seen": 38565964, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 1.2093932628631592, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.1752, | |
| "num_input_tokens_seen": 39144446, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.6265527009963989, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1693, | |
| "num_input_tokens_seen": 39709784, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.1358744204044342, | |
| "eval_runtime": 45.1674, | |
| "eval_samples_per_second": 55.35, | |
| "eval_steps_per_second": 6.93, | |
| "num_input_tokens_seen": 39709784, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.8478279709815979, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.1765, | |
| "num_input_tokens_seen": 40271324, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 1.2263400554656982, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.1813, | |
| "num_input_tokens_seen": 40869346, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 1.0012421607971191, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.1843, | |
| "num_input_tokens_seen": 41439134, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 3.3730311393737793, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.1788, | |
| "num_input_tokens_seen": 42021868, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.48923859000205994, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1729, | |
| "num_input_tokens_seen": 42614112, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.13453309237957, | |
| "eval_runtime": 45.1646, | |
| "eval_samples_per_second": 55.353, | |
| "eval_steps_per_second": 6.93, | |
| "num_input_tokens_seen": 42614112, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.5432136654853821, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.1763, | |
| "num_input_tokens_seen": 43221016, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.0637930631637573, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.1718, | |
| "num_input_tokens_seen": 43792306, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.766243577003479, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.1639, | |
| "num_input_tokens_seen": 44358230, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.6301211714744568, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.1782, | |
| "num_input_tokens_seen": 44912720, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.600902557373047, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1711, | |
| "num_input_tokens_seen": 45471178, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.1343887597322464, | |
| "eval_runtime": 45.1761, | |
| "eval_samples_per_second": 55.339, | |
| "eval_steps_per_second": 6.928, | |
| "num_input_tokens_seen": 45471178, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.8937748670578003, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.1775, | |
| "num_input_tokens_seen": 46048204, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.3030432462692261, | |
| "learning_rate": 9e-06, | |
| "loss": 0.1681, | |
| "num_input_tokens_seen": 46617836, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.7168088555335999, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.1649, | |
| "num_input_tokens_seen": 47193690, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 1.0438125133514404, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1745, | |
| "num_input_tokens_seen": 47771028, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.256290078163147, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1735, | |
| "num_input_tokens_seen": 48355942, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.13401633501052856, | |
| "eval_runtime": 45.2261, | |
| "eval_samples_per_second": 55.278, | |
| "eval_steps_per_second": 6.921, | |
| "num_input_tokens_seen": 48355942, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.8616393804550171, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.1721, | |
| "num_input_tokens_seen": 48935198, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 1.0820131301879883, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.1729, | |
| "num_input_tokens_seen": 49486184, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 1.0676862001419067, | |
| "learning_rate": 6e-06, | |
| "loss": 0.174, | |
| "num_input_tokens_seen": 50057436, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.8204198479652405, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.1741, | |
| "num_input_tokens_seen": 50627550, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.747173011302948, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1797, | |
| "num_input_tokens_seen": 51187066, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.13398794829845428, | |
| "eval_runtime": 45.1424, | |
| "eval_samples_per_second": 55.38, | |
| "eval_steps_per_second": 6.934, | |
| "num_input_tokens_seen": 51187066, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 1.2061275243759155, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.1744, | |
| "num_input_tokens_seen": 51768826, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.8262666463851929, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.1681, | |
| "num_input_tokens_seen": 52332322, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.3594360649585724, | |
| "learning_rate": 3.5000000000000004e-06, | |
| "loss": 0.1653, | |
| "num_input_tokens_seen": 52916462, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.3951094448566437, | |
| "learning_rate": 3e-06, | |
| "loss": 0.1584, | |
| "num_input_tokens_seen": 53487326, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 1.0189286470413208, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1659, | |
| "num_input_tokens_seen": 54074434, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.13384351134300232, | |
| "eval_runtime": 45.2087, | |
| "eval_samples_per_second": 55.299, | |
| "eval_steps_per_second": 6.923, | |
| "num_input_tokens_seen": 54074434, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.1751518249511719, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.1664, | |
| "num_input_tokens_seen": 54651492, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.47343894839286804, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.172, | |
| "num_input_tokens_seen": 55223994, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.70182865858078, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.175, | |
| "num_input_tokens_seen": 55785026, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.8210330605506897, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.1706, | |
| "num_input_tokens_seen": 56363982, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.5335518717765808, | |
| "learning_rate": 0.0, | |
| "loss": 0.1658, | |
| "num_input_tokens_seen": 56941616, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.13376566767692566, | |
| "eval_runtime": 45.1906, | |
| "eval_samples_per_second": 55.321, | |
| "eval_steps_per_second": 6.926, | |
| "num_input_tokens_seen": 56941616, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "num_input_tokens_seen": 56941616, | |
| "step": 50000, | |
| "total_flos": 2.067540416810189e+16, | |
| "train_loss": 0.19179043090820314, | |
| "train_runtime": 4730.669, | |
| "train_samples_per_second": 21.139, | |
| "train_steps_per_second": 10.569, | |
| "train_tokens_per_second": 12036.694 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 50000, | |
| "num_input_tokens_seen": 56941616, | |
| "num_train_epochs": 1, | |
| "save_steps": 2500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.067540416810189e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |