| { | |
| "best_metric": 0.18627458810806274, | |
| "best_model_checkpoint": "/data1/CACHE/huggingface/hg_training/traing_whisper_base_9dataset_multi-gpu4-wnoise/checkpoint-38000", | |
| "epoch": 6.846846846846847, | |
| "eval_steps": 1000, | |
| "global_step": 38000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 6.807682991027832, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.579, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 4.137929916381836, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.6054, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 3.63185977935791, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.4771, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 4.2249555587768555, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.4225, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 4.122411251068115, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.387, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 3.319619655609131, | |
| "learning_rate": 9.982545454545457e-06, | |
| "loss": 0.3653, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 2.9246363639831543, | |
| "learning_rate": 9.964363636363637e-06, | |
| "loss": 0.3418, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 3.4514670372009277, | |
| "learning_rate": 9.946181818181819e-06, | |
| "loss": 0.3389, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 3.152733325958252, | |
| "learning_rate": 9.928e-06, | |
| "loss": 0.3159, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 2.9511702060699463, | |
| "learning_rate": 9.909818181818182e-06, | |
| "loss": 0.3191, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 0.356253445148468, | |
| "eval_runtime": 445.2731, | |
| "eval_samples_per_second": 136.709, | |
| "eval_steps_per_second": 4.274, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.535557508468628, | |
| "learning_rate": 9.891636363636364e-06, | |
| "loss": 0.3103, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.9129879474639893, | |
| "learning_rate": 9.873454545454546e-06, | |
| "loss": 0.3024, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 3.5554208755493164, | |
| "learning_rate": 9.855272727272728e-06, | |
| "loss": 0.3021, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 3.1638295650482178, | |
| "learning_rate": 9.83709090909091e-06, | |
| "loss": 0.2995, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 3.0975465774536133, | |
| "learning_rate": 9.818909090909092e-06, | |
| "loss": 0.3014, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 3.0966498851776123, | |
| "learning_rate": 9.800727272727273e-06, | |
| "loss": 0.2841, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 2.612806558609009, | |
| "learning_rate": 9.782545454545455e-06, | |
| "loss": 0.2821, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 3.115729808807373, | |
| "learning_rate": 9.764363636363637e-06, | |
| "loss": 0.2792, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 2.9033608436584473, | |
| "learning_rate": 9.746181818181819e-06, | |
| "loss": 0.2835, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 2.7176175117492676, | |
| "learning_rate": 9.728e-06, | |
| "loss": 0.2667, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 0.30776357650756836, | |
| "eval_runtime": 447.4332, | |
| "eval_samples_per_second": 136.049, | |
| "eval_steps_per_second": 4.253, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 3.1798152923583984, | |
| "learning_rate": 9.709818181818183e-06, | |
| "loss": 0.2637, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.6578774452209473, | |
| "learning_rate": 9.691636363636364e-06, | |
| "loss": 0.267, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 2.983422040939331, | |
| "learning_rate": 9.673636363636364e-06, | |
| "loss": 0.2674, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 2.7927327156066895, | |
| "learning_rate": 9.655454545454547e-06, | |
| "loss": 0.2575, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 3.017789363861084, | |
| "learning_rate": 9.63727272727273e-06, | |
| "loss": 0.2635, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 3.1174545288085938, | |
| "learning_rate": 9.61909090909091e-06, | |
| "loss": 0.2619, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 3.1417720317840576, | |
| "learning_rate": 9.600909090909091e-06, | |
| "loss": 0.2612, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 2.9511682987213135, | |
| "learning_rate": 9.582727272727273e-06, | |
| "loss": 0.2535, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 2.670490026473999, | |
| "learning_rate": 9.564545454545455e-06, | |
| "loss": 0.2542, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 2.8334949016571045, | |
| "learning_rate": 9.546363636363637e-06, | |
| "loss": 0.2448, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 0.28384605050086975, | |
| "eval_runtime": 363.3283, | |
| "eval_samples_per_second": 167.543, | |
| "eval_steps_per_second": 5.238, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 3.076014757156372, | |
| "learning_rate": 9.528181818181819e-06, | |
| "loss": 0.2497, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 2.6771583557128906, | |
| "learning_rate": 9.51e-06, | |
| "loss": 0.2474, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 2.992527484893799, | |
| "learning_rate": 9.491818181818182e-06, | |
| "loss": 0.2452, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 2.8110687732696533, | |
| "learning_rate": 9.473636363636364e-06, | |
| "loss": 0.2381, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 2.5542688369750977, | |
| "learning_rate": 9.455454545454546e-06, | |
| "loss": 0.2436, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 2.7978878021240234, | |
| "learning_rate": 9.437272727272728e-06, | |
| "loss": 0.2387, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 3.2161848545074463, | |
| "learning_rate": 9.41909090909091e-06, | |
| "loss": 0.2421, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 3.0258097648620605, | |
| "learning_rate": 9.400909090909091e-06, | |
| "loss": 0.2415, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.86924409866333, | |
| "learning_rate": 9.382727272727273e-06, | |
| "loss": 0.2351, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.6659717559814453, | |
| "learning_rate": 9.364545454545455e-06, | |
| "loss": 0.2322, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 0.26763468980789185, | |
| "eval_runtime": 459.6944, | |
| "eval_samples_per_second": 132.421, | |
| "eval_steps_per_second": 4.14, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 2.9035747051239014, | |
| "learning_rate": 9.346363636363637e-06, | |
| "loss": 0.2311, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.607459545135498, | |
| "learning_rate": 9.328181818181819e-06, | |
| "loss": 0.2356, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 3.0524134635925293, | |
| "learning_rate": 9.31e-06, | |
| "loss": 0.2315, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 3.0532467365264893, | |
| "learning_rate": 9.291818181818182e-06, | |
| "loss": 0.2275, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 2.4372193813323975, | |
| "learning_rate": 9.273636363636364e-06, | |
| "loss": 0.2373, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 2.6966984272003174, | |
| "learning_rate": 9.255454545454546e-06, | |
| "loss": 0.2235, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 2.752675771713257, | |
| "learning_rate": 9.237272727272728e-06, | |
| "loss": 0.2298, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 2.880847692489624, | |
| "learning_rate": 9.21909090909091e-06, | |
| "loss": 0.2268, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 2.682913303375244, | |
| "learning_rate": 9.200909090909092e-06, | |
| "loss": 0.2224, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 2.731718063354492, | |
| "learning_rate": 9.182727272727274e-06, | |
| "loss": 0.2171, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.25470873713493347, | |
| "eval_runtime": 486.5055, | |
| "eval_samples_per_second": 125.123, | |
| "eval_steps_per_second": 3.912, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 3.1988534927368164, | |
| "learning_rate": 9.164545454545455e-06, | |
| "loss": 0.2215, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 2.4781415462493896, | |
| "learning_rate": 9.146363636363637e-06, | |
| "loss": 0.217, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 2.7429697513580322, | |
| "learning_rate": 9.128181818181819e-06, | |
| "loss": 0.2196, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 2.868225336074829, | |
| "learning_rate": 9.110000000000001e-06, | |
| "loss": 0.2241, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 2.566040277481079, | |
| "learning_rate": 9.091818181818183e-06, | |
| "loss": 0.2208, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 2.6690568923950195, | |
| "learning_rate": 9.073636363636365e-06, | |
| "loss": 0.2082, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 2.438952684402466, | |
| "learning_rate": 9.055454545454546e-06, | |
| "loss": 0.2009, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 2.5286898612976074, | |
| "learning_rate": 9.037272727272728e-06, | |
| "loss": 0.2081, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 2.2770423889160156, | |
| "learning_rate": 9.01909090909091e-06, | |
| "loss": 0.1966, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 2.502218723297119, | |
| "learning_rate": 9.000909090909092e-06, | |
| "loss": 0.2037, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 0.24770112335681915, | |
| "eval_runtime": 501.2949, | |
| "eval_samples_per_second": 121.432, | |
| "eval_steps_per_second": 3.796, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 2.348766565322876, | |
| "learning_rate": 8.982727272727274e-06, | |
| "loss": 0.1958, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 2.7815017700195312, | |
| "learning_rate": 8.964545454545456e-06, | |
| "loss": 0.1943, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 2.151355266571045, | |
| "learning_rate": 8.946363636363637e-06, | |
| "loss": 0.1911, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 2.5474178791046143, | |
| "learning_rate": 8.92818181818182e-06, | |
| "loss": 0.1988, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 2.5478312969207764, | |
| "learning_rate": 8.910181818181819e-06, | |
| "loss": 0.197, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 2.358614683151245, | |
| "learning_rate": 8.892e-06, | |
| "loss": 0.1918, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 3.1825013160705566, | |
| "learning_rate": 8.873818181818182e-06, | |
| "loss": 0.1976, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 2.506746530532837, | |
| "learning_rate": 8.855636363636364e-06, | |
| "loss": 0.1954, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 2.5737404823303223, | |
| "learning_rate": 8.837454545454546e-06, | |
| "loss": 0.1958, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 3.3842029571533203, | |
| "learning_rate": 8.819272727272728e-06, | |
| "loss": 0.1913, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 0.23902775347232819, | |
| "eval_runtime": 354.6913, | |
| "eval_samples_per_second": 171.622, | |
| "eval_steps_per_second": 5.365, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.24104642868042, | |
| "learning_rate": 8.80109090909091e-06, | |
| "loss": 0.1957, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 2.8391642570495605, | |
| "learning_rate": 8.782909090909092e-06, | |
| "loss": 0.1987, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 2.0376856327056885, | |
| "learning_rate": 8.764727272727273e-06, | |
| "loss": 0.1868, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 3.101025104522705, | |
| "learning_rate": 8.746545454545455e-06, | |
| "loss": 0.188, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 2.6974034309387207, | |
| "learning_rate": 8.728363636363637e-06, | |
| "loss": 0.1945, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 2.4040937423706055, | |
| "learning_rate": 8.710181818181819e-06, | |
| "loss": 0.1914, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 2.8795206546783447, | |
| "learning_rate": 8.692e-06, | |
| "loss": 0.1982, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 2.352360486984253, | |
| "learning_rate": 8.673818181818183e-06, | |
| "loss": 0.1867, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 2.6391663551330566, | |
| "learning_rate": 8.655636363636364e-06, | |
| "loss": 0.1822, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 2.915161609649658, | |
| "learning_rate": 8.637454545454546e-06, | |
| "loss": 0.1903, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_loss": 0.23261629045009613, | |
| "eval_runtime": 492.9612, | |
| "eval_samples_per_second": 123.484, | |
| "eval_steps_per_second": 3.86, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 2.6614203453063965, | |
| "learning_rate": 8.619272727272728e-06, | |
| "loss": 0.1862, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 2.8267104625701904, | |
| "learning_rate": 8.60109090909091e-06, | |
| "loss": 0.1848, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.4803242683410645, | |
| "learning_rate": 8.582909090909092e-06, | |
| "loss": 0.1858, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 3.031543254852295, | |
| "learning_rate": 8.564727272727274e-06, | |
| "loss": 0.1816, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 2.55397629737854, | |
| "learning_rate": 8.546545454545456e-06, | |
| "loss": 0.1896, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 2.7617721557617188, | |
| "learning_rate": 8.528363636363637e-06, | |
| "loss": 0.1823, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 2.09851336479187, | |
| "learning_rate": 8.510363636363637e-06, | |
| "loss": 0.1811, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 2.137387990951538, | |
| "learning_rate": 8.492363636363638e-06, | |
| "loss": 0.183, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 2.270770788192749, | |
| "learning_rate": 8.47418181818182e-06, | |
| "loss": 0.1918, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 2.41756010055542, | |
| "learning_rate": 8.456000000000002e-06, | |
| "loss": 0.1829, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 0.22808903455734253, | |
| "eval_runtime": 666.6262, | |
| "eval_samples_per_second": 91.315, | |
| "eval_steps_per_second": 2.855, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 2.676539421081543, | |
| "learning_rate": 8.437818181818182e-06, | |
| "loss": 0.1869, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 2.636401414871216, | |
| "learning_rate": 8.419636363636364e-06, | |
| "loss": 0.1847, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 3.2688345909118652, | |
| "learning_rate": 8.401454545454546e-06, | |
| "loss": 0.1864, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 2.5359675884246826, | |
| "learning_rate": 8.383272727272727e-06, | |
| "loss": 0.1824, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 2.2516977787017822, | |
| "learning_rate": 8.36509090909091e-06, | |
| "loss": 0.1834, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 3.052272081375122, | |
| "learning_rate": 8.346909090909091e-06, | |
| "loss": 0.1746, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 3.142702579498291, | |
| "learning_rate": 8.328727272727275e-06, | |
| "loss": 0.1887, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 2.60490345954895, | |
| "learning_rate": 8.310545454545456e-06, | |
| "loss": 0.1799, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 2.252636432647705, | |
| "learning_rate": 8.292363636363637e-06, | |
| "loss": 0.1721, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.509241819381714, | |
| "learning_rate": 8.274181818181818e-06, | |
| "loss": 0.1822, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 0.22305500507354736, | |
| "eval_runtime": 699.489, | |
| "eval_samples_per_second": 87.025, | |
| "eval_steps_per_second": 2.721, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 2.6383330821990967, | |
| "learning_rate": 8.256e-06, | |
| "loss": 0.1838, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 2.926187753677368, | |
| "learning_rate": 8.237818181818182e-06, | |
| "loss": 0.1793, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 2.827836036682129, | |
| "learning_rate": 8.219636363636364e-06, | |
| "loss": 0.174, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 2.2314600944519043, | |
| "learning_rate": 8.201454545454546e-06, | |
| "loss": 0.1763, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 2.6388306617736816, | |
| "learning_rate": 8.183272727272728e-06, | |
| "loss": 0.1801, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 2.648263931274414, | |
| "learning_rate": 8.165090909090911e-06, | |
| "loss": 0.1775, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 2.5469634532928467, | |
| "learning_rate": 8.146909090909091e-06, | |
| "loss": 0.1769, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 2.1157877445220947, | |
| "learning_rate": 8.128727272727273e-06, | |
| "loss": 0.1833, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 2.086653709411621, | |
| "learning_rate": 8.110545454545455e-06, | |
| "loss": 0.1787, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 2.8866491317749023, | |
| "learning_rate": 8.092363636363637e-06, | |
| "loss": 0.1729, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 0.2171076089143753, | |
| "eval_runtime": 564.1273, | |
| "eval_samples_per_second": 107.906, | |
| "eval_steps_per_second": 3.373, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.083310604095459, | |
| "learning_rate": 8.074181818181819e-06, | |
| "loss": 0.1774, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 2.184401750564575, | |
| "learning_rate": 8.056e-06, | |
| "loss": 0.1616, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 2.4366555213928223, | |
| "learning_rate": 8.037818181818182e-06, | |
| "loss": 0.1633, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 2.246950626373291, | |
| "learning_rate": 8.019636363636364e-06, | |
| "loss": 0.1623, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 2.4729185104370117, | |
| "learning_rate": 8.001454545454546e-06, | |
| "loss": 0.1544, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 2.3734850883483887, | |
| "learning_rate": 7.983272727272728e-06, | |
| "loss": 0.1595, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 2.2376785278320312, | |
| "learning_rate": 7.96509090909091e-06, | |
| "loss": 0.1561, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 2.2780098915100098, | |
| "learning_rate": 7.946909090909091e-06, | |
| "loss": 0.1549, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 2.2681996822357178, | |
| "learning_rate": 7.928727272727273e-06, | |
| "loss": 0.1641, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 2.3715972900390625, | |
| "learning_rate": 7.910545454545455e-06, | |
| "loss": 0.1611, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "eval_loss": 0.2156703770160675, | |
| "eval_runtime": 477.7675, | |
| "eval_samples_per_second": 127.411, | |
| "eval_steps_per_second": 3.983, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 1.9450706243515015, | |
| "learning_rate": 7.892363636363637e-06, | |
| "loss": 0.152, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 2.151553153991699, | |
| "learning_rate": 7.874181818181819e-06, | |
| "loss": 0.1564, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 2.2513058185577393, | |
| "learning_rate": 7.856e-06, | |
| "loss": 0.154, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 2.370582342147827, | |
| "learning_rate": 7.837818181818183e-06, | |
| "loss": 0.1491, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 2.598097324371338, | |
| "learning_rate": 7.819636363636364e-06, | |
| "loss": 0.1579, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 2.6642096042633057, | |
| "learning_rate": 7.801636363636364e-06, | |
| "loss": 0.1576, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 2.4024887084960938, | |
| "learning_rate": 7.783454545454546e-06, | |
| "loss": 0.1599, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 2.2930634021759033, | |
| "learning_rate": 7.765272727272728e-06, | |
| "loss": 0.1538, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 2.1747636795043945, | |
| "learning_rate": 7.74709090909091e-06, | |
| "loss": 0.1598, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 2.725010395050049, | |
| "learning_rate": 7.728909090909091e-06, | |
| "loss": 0.1509, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_loss": 0.21198728680610657, | |
| "eval_runtime": 596.3747, | |
| "eval_samples_per_second": 102.072, | |
| "eval_steps_per_second": 3.191, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 2.2646334171295166, | |
| "learning_rate": 7.710727272727273e-06, | |
| "loss": 0.1569, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 2.8910107612609863, | |
| "learning_rate": 7.692545454545455e-06, | |
| "loss": 0.161, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.7823307514190674, | |
| "learning_rate": 7.674363636363637e-06, | |
| "loss": 0.1642, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 3.888598918914795, | |
| "learning_rate": 7.656181818181819e-06, | |
| "loss": 0.1565, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 2.2702815532684326, | |
| "learning_rate": 7.638e-06, | |
| "loss": 0.1546, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 2.49238920211792, | |
| "learning_rate": 7.619818181818183e-06, | |
| "loss": 0.1562, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 2.2743749618530273, | |
| "learning_rate": 7.601636363636364e-06, | |
| "loss": 0.1611, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 2.550445318222046, | |
| "learning_rate": 7.583454545454546e-06, | |
| "loss": 0.1569, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.858296513557434, | |
| "learning_rate": 7.565272727272728e-06, | |
| "loss": 0.1642, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 2.714526891708374, | |
| "learning_rate": 7.54709090909091e-06, | |
| "loss": 0.1552, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_loss": 0.20844437181949615, | |
| "eval_runtime": 567.6604, | |
| "eval_samples_per_second": 107.235, | |
| "eval_steps_per_second": 3.352, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 2.246244430541992, | |
| "learning_rate": 7.528909090909091e-06, | |
| "loss": 0.1545, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 2.7097859382629395, | |
| "learning_rate": 7.510727272727273e-06, | |
| "loss": 0.1547, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 3.1044564247131348, | |
| "learning_rate": 7.492545454545456e-06, | |
| "loss": 0.1561, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 3.426424741744995, | |
| "learning_rate": 7.474363636363638e-06, | |
| "loss": 0.1545, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 2.8595056533813477, | |
| "learning_rate": 7.456181818181819e-06, | |
| "loss": 0.157, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 2.094259023666382, | |
| "learning_rate": 7.438000000000001e-06, | |
| "loss": 0.1582, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 2.1589813232421875, | |
| "learning_rate": 7.4198181818181825e-06, | |
| "loss": 0.1608, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 2.2940006256103516, | |
| "learning_rate": 7.401636363636364e-06, | |
| "loss": 0.1468, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 2.0454187393188477, | |
| "learning_rate": 7.383454545454546e-06, | |
| "loss": 0.1535, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 2.362870693206787, | |
| "learning_rate": 7.365272727272728e-06, | |
| "loss": 0.1487, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 0.20696710050106049, | |
| "eval_runtime": 550.539, | |
| "eval_samples_per_second": 110.57, | |
| "eval_steps_per_second": 3.457, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 2.293924331665039, | |
| "learning_rate": 7.347090909090909e-06, | |
| "loss": 0.1565, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 2.045642614364624, | |
| "learning_rate": 7.328909090909091e-06, | |
| "loss": 0.1545, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 2.3518779277801514, | |
| "learning_rate": 7.3107272727272735e-06, | |
| "loss": 0.1581, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "grad_norm": 2.111579418182373, | |
| "learning_rate": 7.292545454545455e-06, | |
| "loss": 0.1578, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 2.008013963699341, | |
| "learning_rate": 7.274363636363637e-06, | |
| "loss": 0.1459, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 2.2898120880126953, | |
| "learning_rate": 7.256181818181819e-06, | |
| "loss": 0.1543, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 2.504873752593994, | |
| "learning_rate": 7.238000000000001e-06, | |
| "loss": 0.1485, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 2.294981002807617, | |
| "learning_rate": 7.219818181818183e-06, | |
| "loss": 0.1542, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 2.4189417362213135, | |
| "learning_rate": 7.201636363636364e-06, | |
| "loss": 0.1523, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 2.0170607566833496, | |
| "learning_rate": 7.1834545454545455e-06, | |
| "loss": 0.1492, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 0.20441067218780518, | |
| "eval_runtime": 568.7903, | |
| "eval_samples_per_second": 107.022, | |
| "eval_steps_per_second": 3.346, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 2.5287556648254395, | |
| "learning_rate": 7.165272727272727e-06, | |
| "loss": 0.1519, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 2.236844301223755, | |
| "learning_rate": 7.1472727272727285e-06, | |
| "loss": 0.1519, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 2.3229405879974365, | |
| "learning_rate": 7.12909090909091e-06, | |
| "loss": 0.1561, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 2.1688249111175537, | |
| "learning_rate": 7.110909090909091e-06, | |
| "loss": 0.1471, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 2.4697909355163574, | |
| "learning_rate": 7.092727272727273e-06, | |
| "loss": 0.1538, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 2.3912057876586914, | |
| "learning_rate": 7.074545454545455e-06, | |
| "loss": 0.1607, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 3.0056750774383545, | |
| "learning_rate": 7.056363636363637e-06, | |
| "loss": 0.1468, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 2.248894691467285, | |
| "learning_rate": 7.038181818181819e-06, | |
| "loss": 0.139, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 2.268486976623535, | |
| "learning_rate": 7.0200000000000006e-06, | |
| "loss": 0.1363, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 1.9803905487060547, | |
| "learning_rate": 7.0018181818181815e-06, | |
| "loss": 0.1317, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "eval_loss": 0.20176592469215393, | |
| "eval_runtime": 557.9491, | |
| "eval_samples_per_second": 109.101, | |
| "eval_steps_per_second": 3.411, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 2.0611746311187744, | |
| "learning_rate": 6.983636363636365e-06, | |
| "loss": 0.135, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 1.9974679946899414, | |
| "learning_rate": 6.965454545454546e-06, | |
| "loss": 0.1393, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 2.247957229614258, | |
| "learning_rate": 6.947272727272728e-06, | |
| "loss": 0.1377, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "grad_norm": 2.3139231204986572, | |
| "learning_rate": 6.92909090909091e-06, | |
| "loss": 0.1402, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 2.253744602203369, | |
| "learning_rate": 6.910909090909092e-06, | |
| "loss": 0.1364, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 2.3517744541168213, | |
| "learning_rate": 6.892727272727273e-06, | |
| "loss": 0.137, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 1.9931228160858154, | |
| "learning_rate": 6.874545454545455e-06, | |
| "loss": 0.1417, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "grad_norm": 1.9905173778533936, | |
| "learning_rate": 6.856363636363636e-06, | |
| "loss": 0.1337, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 2.7830097675323486, | |
| "learning_rate": 6.838181818181818e-06, | |
| "loss": 0.1364, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "grad_norm": 2.7897889614105225, | |
| "learning_rate": 6.820000000000001e-06, | |
| "loss": 0.1361, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_loss": 0.20188076794147491, | |
| "eval_runtime": 572.624, | |
| "eval_samples_per_second": 106.305, | |
| "eval_steps_per_second": 3.323, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "grad_norm": 2.2669436931610107, | |
| "learning_rate": 6.801818181818183e-06, | |
| "loss": 0.1394, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "grad_norm": 1.8945894241333008, | |
| "learning_rate": 6.7836363636363644e-06, | |
| "loss": 0.1358, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 2.3417062759399414, | |
| "learning_rate": 6.765454545454546e-06, | |
| "loss": 0.1369, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 2.118128776550293, | |
| "learning_rate": 6.747272727272728e-06, | |
| "loss": 0.1324, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 2.1800739765167236, | |
| "learning_rate": 6.72909090909091e-06, | |
| "loss": 0.1366, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 2.5400383472442627, | |
| "learning_rate": 6.710909090909091e-06, | |
| "loss": 0.1364, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "grad_norm": 2.051309823989868, | |
| "learning_rate": 6.692727272727273e-06, | |
| "loss": 0.1328, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 2.2389767169952393, | |
| "learning_rate": 6.674545454545455e-06, | |
| "loss": 0.1379, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "grad_norm": 2.198885917663574, | |
| "learning_rate": 6.6563636363636365e-06, | |
| "loss": 0.1405, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 2.476261615753174, | |
| "learning_rate": 6.638181818181819e-06, | |
| "loss": 0.1358, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "eval_loss": 0.19895973801612854, | |
| "eval_runtime": 554.3371, | |
| "eval_samples_per_second": 109.812, | |
| "eval_steps_per_second": 3.433, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 2.0383458137512207, | |
| "learning_rate": 6.620000000000001e-06, | |
| "loss": 0.1382, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "grad_norm": 2.5287790298461914, | |
| "learning_rate": 6.601818181818183e-06, | |
| "loss": 0.1388, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 2.468118667602539, | |
| "learning_rate": 6.583636363636365e-06, | |
| "loss": 0.1387, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 2.280622959136963, | |
| "learning_rate": 6.565454545454546e-06, | |
| "loss": 0.1352, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 2.0358574390411377, | |
| "learning_rate": 6.5472727272727275e-06, | |
| "loss": 0.1374, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "grad_norm": 2.251955986022949, | |
| "learning_rate": 6.529090909090909e-06, | |
| "loss": 0.1411, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 2.215778350830078, | |
| "learning_rate": 6.510909090909091e-06, | |
| "loss": 0.1422, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "grad_norm": 2.3054020404815674, | |
| "learning_rate": 6.492727272727273e-06, | |
| "loss": 0.1352, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "grad_norm": 2.643420934677124, | |
| "learning_rate": 6.474545454545456e-06, | |
| "loss": 0.1357, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 2.40155291557312, | |
| "learning_rate": 6.4563636363636375e-06, | |
| "loss": 0.1382, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "eval_loss": 0.19792823493480682, | |
| "eval_runtime": 559.0516, | |
| "eval_samples_per_second": 108.886, | |
| "eval_steps_per_second": 3.404, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "grad_norm": 2.9484267234802246, | |
| "learning_rate": 6.438181818181819e-06, | |
| "loss": 0.1353, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 2.413797616958618, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.1361, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "grad_norm": 2.0721383094787598, | |
| "learning_rate": 6.402000000000001e-06, | |
| "loss": 0.1338, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 2.4742093086242676, | |
| "learning_rate": 6.384e-06, | |
| "loss": 0.1337, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "grad_norm": 3.042827606201172, | |
| "learning_rate": 6.365818181818182e-06, | |
| "loss": 0.1385, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 2.4437592029571533, | |
| "learning_rate": 6.347636363636365e-06, | |
| "loss": 0.135, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "grad_norm": 1.9226901531219482, | |
| "learning_rate": 6.3294545454545466e-06, | |
| "loss": 0.1322, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 2.4258248805999756, | |
| "learning_rate": 6.311272727272728e-06, | |
| "loss": 0.1385, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "grad_norm": 2.3766396045684814, | |
| "learning_rate": 6.293090909090909e-06, | |
| "loss": 0.1331, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "grad_norm": 2.920332670211792, | |
| "learning_rate": 6.274909090909091e-06, | |
| "loss": 0.1313, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_loss": 0.1950678676366806, | |
| "eval_runtime": 562.0997, | |
| "eval_samples_per_second": 108.296, | |
| "eval_steps_per_second": 3.386, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 2.008997917175293, | |
| "learning_rate": 6.256727272727273e-06, | |
| "loss": 0.1331, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "grad_norm": 2.3787338733673096, | |
| "learning_rate": 6.238545454545455e-06, | |
| "loss": 0.1321, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 2.1688482761383057, | |
| "learning_rate": 6.220363636363637e-06, | |
| "loss": 0.1347, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "grad_norm": 2.0800232887268066, | |
| "learning_rate": 6.202181818181819e-06, | |
| "loss": 0.135, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 2.2381479740142822, | |
| "learning_rate": 6.184e-06, | |
| "loss": 0.1333, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "grad_norm": 2.4402754306793213, | |
| "learning_rate": 6.165818181818183e-06, | |
| "loss": 0.137, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "grad_norm": 2.3774378299713135, | |
| "learning_rate": 6.147636363636364e-06, | |
| "loss": 0.137, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "grad_norm": 2.5666518211364746, | |
| "learning_rate": 6.129454545454546e-06, | |
| "loss": 0.1364, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 2.7530879974365234, | |
| "learning_rate": 6.111272727272728e-06, | |
| "loss": 0.1329, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 2.2446491718292236, | |
| "learning_rate": 6.09309090909091e-06, | |
| "loss": 0.1286, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "eval_loss": 0.19367100298404694, | |
| "eval_runtime": 556.8108, | |
| "eval_samples_per_second": 109.324, | |
| "eval_steps_per_second": 3.418, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "grad_norm": 2.472163200378418, | |
| "learning_rate": 6.0749090909090915e-06, | |
| "loss": 0.135, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.458562135696411, | |
| "learning_rate": 6.056727272727273e-06, | |
| "loss": 0.1343, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "grad_norm": 2.642648696899414, | |
| "learning_rate": 6.038545454545455e-06, | |
| "loss": 0.1256, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 1.9326255321502686, | |
| "learning_rate": 6.020363636363636e-06, | |
| "loss": 0.1178, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "grad_norm": 2.303805112838745, | |
| "learning_rate": 6.002181818181819e-06, | |
| "loss": 0.1262, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "grad_norm": 1.9432106018066406, | |
| "learning_rate": 5.984000000000001e-06, | |
| "loss": 0.1178, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 2.2067601680755615, | |
| "learning_rate": 5.9658181818181825e-06, | |
| "loss": 0.1163, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 1.9173979759216309, | |
| "learning_rate": 5.947636363636364e-06, | |
| "loss": 0.1182, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 1.9207819700241089, | |
| "learning_rate": 5.929454545454546e-06, | |
| "loss": 0.1195, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 2.341498851776123, | |
| "learning_rate": 5.911272727272728e-06, | |
| "loss": 0.1223, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_loss": 0.19433893263339996, | |
| "eval_runtime": 559.9176, | |
| "eval_samples_per_second": 108.718, | |
| "eval_steps_per_second": 3.399, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 1.9997601509094238, | |
| "learning_rate": 5.89309090909091e-06, | |
| "loss": 0.128, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "grad_norm": 1.9088443517684937, | |
| "learning_rate": 5.874909090909091e-06, | |
| "loss": 0.1219, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 2.0948617458343506, | |
| "learning_rate": 5.856909090909091e-06, | |
| "loss": 0.1247, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "grad_norm": 2.289156436920166, | |
| "learning_rate": 5.838727272727274e-06, | |
| "loss": 0.1207, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "grad_norm": 3.1126627922058105, | |
| "learning_rate": 5.820545454545456e-06, | |
| "loss": 0.1211, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 2.562894582748413, | |
| "learning_rate": 5.802363636363637e-06, | |
| "loss": 0.1216, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "grad_norm": 2.3009512424468994, | |
| "learning_rate": 5.7841818181818185e-06, | |
| "loss": 0.1228, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "grad_norm": 2.5779926776885986, | |
| "learning_rate": 5.766e-06, | |
| "loss": 0.1214, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "grad_norm": 2.466285228729248, | |
| "learning_rate": 5.747818181818182e-06, | |
| "loss": 0.1206, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 2.0055696964263916, | |
| "learning_rate": 5.729636363636364e-06, | |
| "loss": 0.1256, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "eval_loss": 0.1951528936624527, | |
| "eval_runtime": 566.0319, | |
| "eval_samples_per_second": 107.543, | |
| "eval_steps_per_second": 3.362, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "grad_norm": 1.853898525238037, | |
| "learning_rate": 5.711454545454546e-06, | |
| "loss": 0.121, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "grad_norm": 2.03159499168396, | |
| "learning_rate": 5.693272727272727e-06, | |
| "loss": 0.1252, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "grad_norm": 2.1461386680603027, | |
| "learning_rate": 5.67509090909091e-06, | |
| "loss": 0.1165, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 2.047924041748047, | |
| "learning_rate": 5.656909090909091e-06, | |
| "loss": 0.127, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "grad_norm": 2.1523540019989014, | |
| "learning_rate": 5.638727272727273e-06, | |
| "loss": 0.1201, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "grad_norm": 1.8761101961135864, | |
| "learning_rate": 5.620545454545455e-06, | |
| "loss": 0.1204, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "grad_norm": 1.9362976551055908, | |
| "learning_rate": 5.602363636363637e-06, | |
| "loss": 0.1163, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "grad_norm": 2.4700512886047363, | |
| "learning_rate": 5.584181818181819e-06, | |
| "loss": 0.1177, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 2.274169683456421, | |
| "learning_rate": 5.566000000000001e-06, | |
| "loss": 0.125, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 2.2185416221618652, | |
| "learning_rate": 5.5478181818181816e-06, | |
| "loss": 0.1216, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_loss": 0.19246701896190643, | |
| "eval_runtime": 562.9559, | |
| "eval_samples_per_second": 108.131, | |
| "eval_steps_per_second": 3.38, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 1.9934074878692627, | |
| "learning_rate": 5.529636363636363e-06, | |
| "loss": 0.1204, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "grad_norm": 1.9352362155914307, | |
| "learning_rate": 5.511454545454545e-06, | |
| "loss": 0.1221, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "grad_norm": 2.508136034011841, | |
| "learning_rate": 5.493272727272728e-06, | |
| "loss": 0.1243, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "grad_norm": 1.9421477317810059, | |
| "learning_rate": 5.47509090909091e-06, | |
| "loss": 0.1257, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "grad_norm": 2.0170023441314697, | |
| "learning_rate": 5.456909090909092e-06, | |
| "loss": 0.1213, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "grad_norm": 2.4295244216918945, | |
| "learning_rate": 5.438909090909091e-06, | |
| "loss": 0.1285, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "grad_norm": 2.2002458572387695, | |
| "learning_rate": 5.420727272727273e-06, | |
| "loss": 0.1216, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 2.3806753158569336, | |
| "learning_rate": 5.402545454545455e-06, | |
| "loss": 0.1283, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "grad_norm": 2.4009785652160645, | |
| "learning_rate": 5.384363636363637e-06, | |
| "loss": 0.1172, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 1.9371693134307861, | |
| "learning_rate": 5.3661818181818185e-06, | |
| "loss": 0.1204, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "eval_loss": 0.19222472608089447, | |
| "eval_runtime": 560.4905, | |
| "eval_samples_per_second": 108.607, | |
| "eval_steps_per_second": 3.395, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "grad_norm": 2.303044557571411, | |
| "learning_rate": 5.348000000000001e-06, | |
| "loss": 0.1242, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 2.349740743637085, | |
| "learning_rate": 5.329818181818183e-06, | |
| "loss": 0.1253, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "grad_norm": 2.497861623764038, | |
| "learning_rate": 5.311636363636364e-06, | |
| "loss": 0.1231, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "grad_norm": 2.0001113414764404, | |
| "learning_rate": 5.293454545454546e-06, | |
| "loss": 0.1237, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "grad_norm": 2.0054848194122314, | |
| "learning_rate": 5.275272727272728e-06, | |
| "loss": 0.1169, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "grad_norm": 2.3605902194976807, | |
| "learning_rate": 5.2570909090909095e-06, | |
| "loss": 0.1229, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "grad_norm": 2.1442110538482666, | |
| "learning_rate": 5.238909090909091e-06, | |
| "loss": 0.1186, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "grad_norm": 1.9753350019454956, | |
| "learning_rate": 5.220727272727273e-06, | |
| "loss": 0.1207, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "grad_norm": 2.507814884185791, | |
| "learning_rate": 5.202545454545454e-06, | |
| "loss": 0.1239, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "grad_norm": 2.083677291870117, | |
| "learning_rate": 5.184363636363636e-06, | |
| "loss": 0.125, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "eval_loss": 0.18878485262393951, | |
| "eval_runtime": 570.8746, | |
| "eval_samples_per_second": 106.631, | |
| "eval_steps_per_second": 3.333, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "grad_norm": 2.1022801399230957, | |
| "learning_rate": 5.166181818181819e-06, | |
| "loss": 0.1175, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "grad_norm": 2.282572031021118, | |
| "learning_rate": 5.1480000000000005e-06, | |
| "loss": 0.1209, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "grad_norm": 2.1377222537994385, | |
| "learning_rate": 5.129818181818182e-06, | |
| "loss": 0.12, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "grad_norm": 2.226515769958496, | |
| "learning_rate": 5.111636363636364e-06, | |
| "loss": 0.1183, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "grad_norm": 1.9465168714523315, | |
| "learning_rate": 5.093454545454546e-06, | |
| "loss": 0.1197, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "grad_norm": 2.625356912612915, | |
| "learning_rate": 5.075272727272728e-06, | |
| "loss": 0.1202, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "grad_norm": 2.350402355194092, | |
| "learning_rate": 5.057090909090909e-06, | |
| "loss": 0.12, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 2.0888664722442627, | |
| "learning_rate": 5.038909090909091e-06, | |
| "loss": 0.111, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "grad_norm": 2.536491632461548, | |
| "learning_rate": 5.0207272727272725e-06, | |
| "loss": 0.108, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "grad_norm": 1.663758397102356, | |
| "learning_rate": 5.002545454545455e-06, | |
| "loss": 0.1093, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "eval_loss": 0.1907467395067215, | |
| "eval_runtime": 558.5701, | |
| "eval_samples_per_second": 108.98, | |
| "eval_steps_per_second": 3.407, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "grad_norm": 2.2518911361694336, | |
| "learning_rate": 4.984363636363636e-06, | |
| "loss": 0.1093, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "grad_norm": 1.9912610054016113, | |
| "learning_rate": 4.966181818181818e-06, | |
| "loss": 0.1102, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 2.848151922225952, | |
| "learning_rate": 4.948000000000001e-06, | |
| "loss": 0.114, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "grad_norm": 2.5407612323760986, | |
| "learning_rate": 4.9298181818181826e-06, | |
| "loss": 0.1157, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "grad_norm": 1.966389536857605, | |
| "learning_rate": 4.9116363636363636e-06, | |
| "loss": 0.1108, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "grad_norm": 2.2150988578796387, | |
| "learning_rate": 4.893454545454545e-06, | |
| "loss": 0.1097, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "grad_norm": 2.6871962547302246, | |
| "learning_rate": 4.875272727272728e-06, | |
| "loss": 0.1078, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "grad_norm": 2.092545986175537, | |
| "learning_rate": 4.85709090909091e-06, | |
| "loss": 0.1139, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "grad_norm": 2.1619746685028076, | |
| "learning_rate": 4.838909090909091e-06, | |
| "loss": 0.1137, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "grad_norm": 2.059086322784424, | |
| "learning_rate": 4.820727272727273e-06, | |
| "loss": 0.1092, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "eval_loss": 0.19213946163654327, | |
| "eval_runtime": 559.5411, | |
| "eval_samples_per_second": 108.791, | |
| "eval_steps_per_second": 3.401, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "grad_norm": 2.200467109680176, | |
| "learning_rate": 4.802545454545455e-06, | |
| "loss": 0.1166, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "grad_norm": 2.1066653728485107, | |
| "learning_rate": 4.784363636363637e-06, | |
| "loss": 0.1116, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 1.9387317895889282, | |
| "learning_rate": 4.766181818181818e-06, | |
| "loss": 0.1158, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "grad_norm": 2.8568620681762695, | |
| "learning_rate": 4.748e-06, | |
| "loss": 0.1101, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "grad_norm": 2.667982816696167, | |
| "learning_rate": 4.729818181818182e-06, | |
| "loss": 0.1154, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "grad_norm": 1.8243011236190796, | |
| "learning_rate": 4.711636363636364e-06, | |
| "loss": 0.1115, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "grad_norm": 2.2636425495147705, | |
| "learning_rate": 4.693636363636364e-06, | |
| "loss": 0.1107, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "grad_norm": 2.183295488357544, | |
| "learning_rate": 4.675454545454546e-06, | |
| "loss": 0.1097, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "grad_norm": 1.9221436977386475, | |
| "learning_rate": 4.657272727272728e-06, | |
| "loss": 0.1102, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "grad_norm": 2.4164745807647705, | |
| "learning_rate": 4.639090909090909e-06, | |
| "loss": 0.1113, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "eval_loss": 0.1900329291820526, | |
| "eval_runtime": 555.8487, | |
| "eval_samples_per_second": 109.514, | |
| "eval_steps_per_second": 3.424, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "grad_norm": 1.9355889558792114, | |
| "learning_rate": 4.6209090909090915e-06, | |
| "loss": 0.1093, | |
| "step": 30100 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "grad_norm": 2.172149419784546, | |
| "learning_rate": 4.602727272727273e-06, | |
| "loss": 0.1091, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "grad_norm": 2.230680465698242, | |
| "learning_rate": 4.584545454545455e-06, | |
| "loss": 0.1068, | |
| "step": 30300 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "grad_norm": 2.4593875408172607, | |
| "learning_rate": 4.566363636363636e-06, | |
| "loss": 0.1076, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 2.358771324157715, | |
| "learning_rate": 4.548181818181819e-06, | |
| "loss": 0.1104, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "grad_norm": 2.006244421005249, | |
| "learning_rate": 4.530000000000001e-06, | |
| "loss": 0.1081, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "grad_norm": 2.121628999710083, | |
| "learning_rate": 4.5118181818181825e-06, | |
| "loss": 0.1152, | |
| "step": 30700 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "grad_norm": 2.2535011768341064, | |
| "learning_rate": 4.4936363636363635e-06, | |
| "loss": 0.113, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "grad_norm": 2.598020553588867, | |
| "learning_rate": 4.475454545454545e-06, | |
| "loss": 0.1082, | |
| "step": 30900 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "grad_norm": 2.041231155395508, | |
| "learning_rate": 4.457272727272728e-06, | |
| "loss": 0.1128, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "eval_loss": 0.1888962835073471, | |
| "eval_runtime": 561.6459, | |
| "eval_samples_per_second": 108.383, | |
| "eval_steps_per_second": 3.388, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 1.9678025245666504, | |
| "learning_rate": 4.43909090909091e-06, | |
| "loss": 0.1049, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "grad_norm": 2.5535237789154053, | |
| "learning_rate": 4.420909090909091e-06, | |
| "loss": 0.1124, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "grad_norm": 2.313497304916382, | |
| "learning_rate": 4.402727272727273e-06, | |
| "loss": 0.111, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "grad_norm": 2.2810420989990234, | |
| "learning_rate": 4.3845454545454545e-06, | |
| "loss": 0.1078, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "grad_norm": 1.816409945487976, | |
| "learning_rate": 4.366363636363637e-06, | |
| "loss": 0.1091, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "grad_norm": 2.5376205444335938, | |
| "learning_rate": 4.348181818181818e-06, | |
| "loss": 0.1124, | |
| "step": 31600 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "grad_norm": 1.8754093647003174, | |
| "learning_rate": 4.33e-06, | |
| "loss": 0.108, | |
| "step": 31700 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "grad_norm": 2.0413951873779297, | |
| "learning_rate": 4.311818181818182e-06, | |
| "loss": 0.1051, | |
| "step": 31800 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 1.9775103330612183, | |
| "learning_rate": 4.293636363636364e-06, | |
| "loss": 0.1101, | |
| "step": 31900 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "grad_norm": 2.9266469478607178, | |
| "learning_rate": 4.2754545454545456e-06, | |
| "loss": 0.1101, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "eval_loss": 0.1881016492843628, | |
| "eval_runtime": 557.9801, | |
| "eval_samples_per_second": 109.095, | |
| "eval_steps_per_second": 3.411, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "grad_norm": 2.3823201656341553, | |
| "learning_rate": 4.257272727272727e-06, | |
| "loss": 0.1103, | |
| "step": 32100 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "grad_norm": 2.3224339485168457, | |
| "learning_rate": 4.239090909090909e-06, | |
| "loss": 0.1117, | |
| "step": 32200 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "grad_norm": 2.2235219478607178, | |
| "learning_rate": 4.220909090909091e-06, | |
| "loss": 0.1093, | |
| "step": 32300 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "grad_norm": 2.0410640239715576, | |
| "learning_rate": 4.202727272727273e-06, | |
| "loss": 0.1099, | |
| "step": 32400 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "grad_norm": 2.325864553451538, | |
| "learning_rate": 4.184545454545455e-06, | |
| "loss": 0.1075, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "grad_norm": 1.8241126537322998, | |
| "learning_rate": 4.166545454545455e-06, | |
| "loss": 0.1082, | |
| "step": 32600 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "grad_norm": 1.5952904224395752, | |
| "learning_rate": 4.148363636363636e-06, | |
| "loss": 0.1108, | |
| "step": 32700 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "grad_norm": 2.1724677085876465, | |
| "learning_rate": 4.130181818181819e-06, | |
| "loss": 0.1094, | |
| "step": 32800 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "grad_norm": 2.2764976024627686, | |
| "learning_rate": 4.112000000000001e-06, | |
| "loss": 0.1044, | |
| "step": 32900 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "grad_norm": 2.123507261276245, | |
| "learning_rate": 4.0938181818181824e-06, | |
| "loss": 0.1083, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "eval_loss": 0.18839485943317413, | |
| "eval_runtime": 556.9779, | |
| "eval_samples_per_second": 109.292, | |
| "eval_steps_per_second": 3.417, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 2.797337532043457, | |
| "learning_rate": 4.0756363636363634e-06, | |
| "loss": 0.1153, | |
| "step": 33100 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "grad_norm": 2.4354584217071533, | |
| "learning_rate": 4.057454545454545e-06, | |
| "loss": 0.1112, | |
| "step": 33200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.2812533378601074, | |
| "learning_rate": 4.039272727272728e-06, | |
| "loss": 0.1068, | |
| "step": 33300 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 1.898974895477295, | |
| "learning_rate": 4.02109090909091e-06, | |
| "loss": 0.099, | |
| "step": 33400 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 2.096282482147217, | |
| "learning_rate": 4.002909090909091e-06, | |
| "loss": 0.1033, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "grad_norm": 2.209646224975586, | |
| "learning_rate": 3.984727272727273e-06, | |
| "loss": 0.1001, | |
| "step": 33600 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 2.0665271282196045, | |
| "learning_rate": 3.966545454545455e-06, | |
| "loss": 0.1007, | |
| "step": 33700 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 2.2653417587280273, | |
| "learning_rate": 3.948363636363637e-06, | |
| "loss": 0.0975, | |
| "step": 33800 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "grad_norm": 1.7271119356155396, | |
| "learning_rate": 3.930181818181818e-06, | |
| "loss": 0.1063, | |
| "step": 33900 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "grad_norm": 1.858734369277954, | |
| "learning_rate": 3.912e-06, | |
| "loss": 0.0983, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "eval_loss": 0.18823260068893433, | |
| "eval_runtime": 563.2389, | |
| "eval_samples_per_second": 108.077, | |
| "eval_steps_per_second": 3.379, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "grad_norm": 2.122073173522949, | |
| "learning_rate": 3.893818181818182e-06, | |
| "loss": 0.1017, | |
| "step": 34100 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "grad_norm": 1.7919000387191772, | |
| "learning_rate": 3.8756363636363645e-06, | |
| "loss": 0.1018, | |
| "step": 34200 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "grad_norm": 1.944100022315979, | |
| "learning_rate": 3.8574545454545455e-06, | |
| "loss": 0.1019, | |
| "step": 34300 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "grad_norm": 2.422239065170288, | |
| "learning_rate": 3.839272727272727e-06, | |
| "loss": 0.1011, | |
| "step": 34400 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "grad_norm": 2.4203903675079346, | |
| "learning_rate": 3.821090909090909e-06, | |
| "loss": 0.1015, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "grad_norm": 2.3504583835601807, | |
| "learning_rate": 3.802909090909091e-06, | |
| "loss": 0.1019, | |
| "step": 34600 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 2.062124729156494, | |
| "learning_rate": 3.7849090909090914e-06, | |
| "loss": 0.1048, | |
| "step": 34700 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "grad_norm": 2.1046996116638184, | |
| "learning_rate": 3.766727272727273e-06, | |
| "loss": 0.1012, | |
| "step": 34800 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "grad_norm": 2.111078977584839, | |
| "learning_rate": 3.7485454545454546e-06, | |
| "loss": 0.1031, | |
| "step": 34900 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "grad_norm": 1.7998141050338745, | |
| "learning_rate": 3.7303636363636364e-06, | |
| "loss": 0.1041, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "eval_loss": 0.1882905513048172, | |
| "eval_runtime": 557.868, | |
| "eval_samples_per_second": 109.117, | |
| "eval_steps_per_second": 3.411, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "grad_norm": 2.569345712661743, | |
| "learning_rate": 3.7121818181818187e-06, | |
| "loss": 0.1015, | |
| "step": 35100 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "grad_norm": 2.156580686569214, | |
| "learning_rate": 3.6940000000000005e-06, | |
| "loss": 0.1007, | |
| "step": 35200 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "grad_norm": 2.1598432064056396, | |
| "learning_rate": 3.675818181818182e-06, | |
| "loss": 0.1035, | |
| "step": 35300 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "grad_norm": 2.94124698638916, | |
| "learning_rate": 3.657636363636364e-06, | |
| "loss": 0.1028, | |
| "step": 35400 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 2.118029832839966, | |
| "learning_rate": 3.639454545454546e-06, | |
| "loss": 0.1027, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "grad_norm": 2.3655309677124023, | |
| "learning_rate": 3.621272727272728e-06, | |
| "loss": 0.1013, | |
| "step": 35600 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "grad_norm": 2.1393494606018066, | |
| "learning_rate": 3.6030909090909093e-06, | |
| "loss": 0.1007, | |
| "step": 35700 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "grad_norm": 2.1543033123016357, | |
| "learning_rate": 3.584909090909091e-06, | |
| "loss": 0.1048, | |
| "step": 35800 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "grad_norm": 2.0389814376831055, | |
| "learning_rate": 3.566727272727273e-06, | |
| "loss": 0.1027, | |
| "step": 35900 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "grad_norm": 1.9840948581695557, | |
| "learning_rate": 3.5485454545454553e-06, | |
| "loss": 0.0997, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "eval_loss": 0.18674355745315552, | |
| "eval_runtime": 559.1771, | |
| "eval_samples_per_second": 108.862, | |
| "eval_steps_per_second": 3.403, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 2.0327794551849365, | |
| "learning_rate": 3.5303636363636367e-06, | |
| "loss": 0.1082, | |
| "step": 36100 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "grad_norm": 1.9940663576126099, | |
| "learning_rate": 3.5121818181818185e-06, | |
| "loss": 0.1008, | |
| "step": 36200 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "grad_norm": 2.4469642639160156, | |
| "learning_rate": 3.4940000000000003e-06, | |
| "loss": 0.0994, | |
| "step": 36300 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "grad_norm": 2.186110258102417, | |
| "learning_rate": 3.4758181818181818e-06, | |
| "loss": 0.0991, | |
| "step": 36400 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "grad_norm": 1.8428528308868408, | |
| "learning_rate": 3.457636363636364e-06, | |
| "loss": 0.1004, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "grad_norm": 2.029137372970581, | |
| "learning_rate": 3.439454545454546e-06, | |
| "loss": 0.1017, | |
| "step": 36600 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "grad_norm": 2.730164051055908, | |
| "learning_rate": 3.4212727272727277e-06, | |
| "loss": 0.105, | |
| "step": 36700 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "grad_norm": 2.557441473007202, | |
| "learning_rate": 3.403090909090909e-06, | |
| "loss": 0.1, | |
| "step": 36800 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "grad_norm": 1.9623521566390991, | |
| "learning_rate": 3.3850909090909095e-06, | |
| "loss": 0.1023, | |
| "step": 36900 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "grad_norm": 2.96305513381958, | |
| "learning_rate": 3.3669090909090913e-06, | |
| "loss": 0.1043, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "eval_loss": 0.18738530576229095, | |
| "eval_runtime": 561.774, | |
| "eval_samples_per_second": 108.359, | |
| "eval_steps_per_second": 3.387, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "grad_norm": 2.373506784439087, | |
| "learning_rate": 3.348727272727273e-06, | |
| "loss": 0.1026, | |
| "step": 37100 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "grad_norm": 2.3987481594085693, | |
| "learning_rate": 3.3305454545454545e-06, | |
| "loss": 0.0965, | |
| "step": 37200 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "grad_norm": 2.418612003326416, | |
| "learning_rate": 3.312363636363637e-06, | |
| "loss": 0.1033, | |
| "step": 37300 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "grad_norm": 2.2459537982940674, | |
| "learning_rate": 3.2941818181818186e-06, | |
| "loss": 0.0968, | |
| "step": 37400 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "grad_norm": 2.8029379844665527, | |
| "learning_rate": 3.2760000000000005e-06, | |
| "loss": 0.1077, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "grad_norm": 2.472376823425293, | |
| "learning_rate": 3.257818181818182e-06, | |
| "loss": 0.1055, | |
| "step": 37600 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "grad_norm": 2.410263776779175, | |
| "learning_rate": 3.2396363636363637e-06, | |
| "loss": 0.0975, | |
| "step": 37700 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "grad_norm": 2.254673719406128, | |
| "learning_rate": 3.221454545454546e-06, | |
| "loss": 0.0997, | |
| "step": 37800 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "grad_norm": 2.2963709831237793, | |
| "learning_rate": 3.203272727272728e-06, | |
| "loss": 0.0991, | |
| "step": 37900 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "grad_norm": 2.1210105419158936, | |
| "learning_rate": 3.1850909090909093e-06, | |
| "loss": 0.0988, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_loss": 0.18627458810806274, | |
| "eval_runtime": 562.6423, | |
| "eval_samples_per_second": 108.191, | |
| "eval_steps_per_second": 3.382, | |
| "step": 38000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 55500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "total_flos": 1.577395142823143e+20, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |