| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.762511493497964, | |
| "eval_steps": 500, | |
| "global_step": 30000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.052541704978326546, | |
| "grad_norm": 0.744549572467804, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 1.663, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10508340995665309, | |
| "grad_norm": 0.6462538838386536, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.616, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15762511493497963, | |
| "grad_norm": 0.5421179533004761, | |
| "learning_rate": 5e-06, | |
| "loss": 1.4549, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.21016681991330619, | |
| "grad_norm": 0.8848033547401428, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.3941, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2627085248916327, | |
| "grad_norm": 0.7000551819801331, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.2752, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2627085248916327, | |
| "eval_loss": 1.0633598566055298, | |
| "eval_runtime": 1.3002, | |
| "eval_samples_per_second": 7.691, | |
| "eval_steps_per_second": 3.845, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31525022986995926, | |
| "grad_norm": 0.8002354502677917, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2205, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3677919348482858, | |
| "grad_norm": 0.7082624435424805, | |
| "learning_rate": 1.1666666666666668e-05, | |
| "loss": 1.188, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.42033363982661237, | |
| "grad_norm": 0.626582682132721, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 1.1647, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.4728753448049389, | |
| "grad_norm": 1.0509939193725586, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.1427, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.5254170497832654, | |
| "grad_norm": 1.1202670335769653, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.144, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5254170497832654, | |
| "eval_loss": 0.9546993374824524, | |
| "eval_runtime": 1.2391, | |
| "eval_samples_per_second": 8.07, | |
| "eval_steps_per_second": 4.035, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.577958754761592, | |
| "grad_norm": 1.1231364011764526, | |
| "learning_rate": 1.8333333333333333e-05, | |
| "loss": 1.0935, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.6305004597399185, | |
| "grad_norm": 1.0733870267868042, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1038, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.6830421647182451, | |
| "grad_norm": 1.0306121110916138, | |
| "learning_rate": 2.1666666666666667e-05, | |
| "loss": 1.1055, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.7355838696965716, | |
| "grad_norm": 1.2776762247085571, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 1.0724, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.7881255746748982, | |
| "grad_norm": 1.0678656101226807, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.0564, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7881255746748982, | |
| "eval_loss": 0.912558913230896, | |
| "eval_runtime": 1.2576, | |
| "eval_samples_per_second": 7.952, | |
| "eval_steps_per_second": 3.976, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8406672796532247, | |
| "grad_norm": 1.6063746213912964, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 1.0391, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.8932089846315513, | |
| "grad_norm": 1.367933750152588, | |
| "learning_rate": 2.8333333333333335e-05, | |
| "loss": 1.0435, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.9457506896098778, | |
| "grad_norm": 1.0610206127166748, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0436, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.9982923945882044, | |
| "grad_norm": 0.8625739216804504, | |
| "learning_rate": 3.1666666666666666e-05, | |
| "loss": 1.0334, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.0508340995665308, | |
| "grad_norm": 1.4904906749725342, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.0228, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0508340995665308, | |
| "eval_loss": 0.8912132978439331, | |
| "eval_runtime": 1.2665, | |
| "eval_samples_per_second": 7.896, | |
| "eval_steps_per_second": 3.948, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.1033758045448574, | |
| "grad_norm": 1.4299525022506714, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.0009, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.155917509523184, | |
| "grad_norm": 1.2776519060134888, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.9988, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.2084592145015105, | |
| "grad_norm": 1.1559321880340576, | |
| "learning_rate": 3.8333333333333334e-05, | |
| "loss": 1.0003, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.261000919479837, | |
| "grad_norm": 0.9451723098754883, | |
| "learning_rate": 4e-05, | |
| "loss": 0.9961, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.3135426244581636, | |
| "grad_norm": 1.4040967226028442, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.9981, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.3135426244581636, | |
| "eval_loss": 0.8769587278366089, | |
| "eval_runtime": 1.2527, | |
| "eval_samples_per_second": 7.983, | |
| "eval_steps_per_second": 3.991, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.3660843294364902, | |
| "grad_norm": 1.469843864440918, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.9868, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.4186260344148167, | |
| "grad_norm": 1.1479499340057373, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.9822, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.4711677393931433, | |
| "grad_norm": 2.1333274841308594, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.9921, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.5237094443714698, | |
| "grad_norm": 1.162014365196228, | |
| "learning_rate": 4.8333333333333334e-05, | |
| "loss": 0.9813, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.5762511493497964, | |
| "grad_norm": 0.9883254170417786, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9796, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5762511493497964, | |
| "eval_loss": 0.8728264570236206, | |
| "eval_runtime": 1.262, | |
| "eval_samples_per_second": 7.924, | |
| "eval_steps_per_second": 3.962, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.628792854328123, | |
| "grad_norm": 1.1366589069366455, | |
| "learning_rate": 4.981481481481482e-05, | |
| "loss": 0.985, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.6813345593064495, | |
| "grad_norm": 1.0342519283294678, | |
| "learning_rate": 4.962962962962963e-05, | |
| "loss": 0.9872, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.733876264284776, | |
| "grad_norm": 0.9889224767684937, | |
| "learning_rate": 4.9444444444444446e-05, | |
| "loss": 0.9892, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.7864179692631026, | |
| "grad_norm": 0.9875237345695496, | |
| "learning_rate": 4.925925925925926e-05, | |
| "loss": 0.9606, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.8389596742414291, | |
| "grad_norm": 0.9079448580741882, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 0.9846, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.8389596742414291, | |
| "eval_loss": 0.869242787361145, | |
| "eval_runtime": 1.2422, | |
| "eval_samples_per_second": 8.05, | |
| "eval_steps_per_second": 4.025, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.8915013792197557, | |
| "grad_norm": 1.0478876829147339, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 0.9616, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.9440430841980822, | |
| "grad_norm": 1.0536926984786987, | |
| "learning_rate": 4.8703703703703704e-05, | |
| "loss": 0.9674, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.9965847891764088, | |
| "grad_norm": 1.1787534952163696, | |
| "learning_rate": 4.851851851851852e-05, | |
| "loss": 0.9567, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.049126494154735, | |
| "grad_norm": 1.0072757005691528, | |
| "learning_rate": 4.8333333333333334e-05, | |
| "loss": 0.9441, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.1016681991330617, | |
| "grad_norm": 1.2437305450439453, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.9316, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.1016681991330617, | |
| "eval_loss": 0.8525739908218384, | |
| "eval_runtime": 1.2391, | |
| "eval_samples_per_second": 8.07, | |
| "eval_steps_per_second": 4.035, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.1542099041113882, | |
| "grad_norm": 1.722901701927185, | |
| "learning_rate": 4.796296296296296e-05, | |
| "loss": 0.934, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.206751609089715, | |
| "grad_norm": 1.032166600227356, | |
| "learning_rate": 4.7777777777777784e-05, | |
| "loss": 0.9324, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.2592933140680413, | |
| "grad_norm": 1.0329689979553223, | |
| "learning_rate": 4.759259259259259e-05, | |
| "loss": 0.9327, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.311835019046368, | |
| "grad_norm": 0.8714644908905029, | |
| "learning_rate": 4.740740740740741e-05, | |
| "loss": 0.9228, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.3643767240246945, | |
| "grad_norm": 1.048893928527832, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.9356, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.3643767240246945, | |
| "eval_loss": 0.8477146029472351, | |
| "eval_runtime": 1.2917, | |
| "eval_samples_per_second": 7.742, | |
| "eval_steps_per_second": 3.871, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.416918429003021, | |
| "grad_norm": 1.1038933992385864, | |
| "learning_rate": 4.703703703703704e-05, | |
| "loss": 0.9317, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.4694601339813476, | |
| "grad_norm": 0.9085045456886292, | |
| "learning_rate": 4.685185185185185e-05, | |
| "loss": 0.9411, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.522001838959674, | |
| "grad_norm": 1.1415351629257202, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.9227, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.5745435439380007, | |
| "grad_norm": 0.9230244755744934, | |
| "learning_rate": 4.648148148148148e-05, | |
| "loss": 0.9362, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.627085248916327, | |
| "grad_norm": 0.9721058011054993, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.9297, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.627085248916327, | |
| "eval_loss": 0.8504465818405151, | |
| "eval_runtime": 1.2834, | |
| "eval_samples_per_second": 7.792, | |
| "eval_steps_per_second": 3.896, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.6796269538946538, | |
| "grad_norm": 0.933588981628418, | |
| "learning_rate": 4.6111111111111115e-05, | |
| "loss": 0.9339, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 2.7321686588729803, | |
| "grad_norm": 0.915013313293457, | |
| "learning_rate": 4.592592592592593e-05, | |
| "loss": 0.9322, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.784710363851307, | |
| "grad_norm": 1.2846601009368896, | |
| "learning_rate": 4.5740740740740745e-05, | |
| "loss": 0.913, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 2.8372520688296334, | |
| "grad_norm": 1.1330525875091553, | |
| "learning_rate": 4.555555555555556e-05, | |
| "loss": 0.9241, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.88979377380796, | |
| "grad_norm": 1.6967169046401978, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 0.9127, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.88979377380796, | |
| "eval_loss": 0.8335689306259155, | |
| "eval_runtime": 1.2838, | |
| "eval_samples_per_second": 7.789, | |
| "eval_steps_per_second": 3.895, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.9423354787862865, | |
| "grad_norm": 0.9576010704040527, | |
| "learning_rate": 4.518518518518519e-05, | |
| "loss": 0.9097, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.994877183764613, | |
| "grad_norm": 1.229671835899353, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.9165, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 3.0474188887429396, | |
| "grad_norm": 0.9229328632354736, | |
| "learning_rate": 4.481481481481482e-05, | |
| "loss": 0.9066, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.099960593721266, | |
| "grad_norm": 1.1152088642120361, | |
| "learning_rate": 4.462962962962963e-05, | |
| "loss": 0.8888, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 3.1525022986995928, | |
| "grad_norm": 1.0948377847671509, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.9233, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.1525022986995928, | |
| "eval_loss": 0.8405929803848267, | |
| "eval_runtime": 1.2669, | |
| "eval_samples_per_second": 7.893, | |
| "eval_steps_per_second": 3.947, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.2050440036779193, | |
| "grad_norm": 1.320657730102539, | |
| "learning_rate": 4.425925925925926e-05, | |
| "loss": 0.9024, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 3.257585708656246, | |
| "grad_norm": 1.3515260219573975, | |
| "learning_rate": 4.4074074074074076e-05, | |
| "loss": 0.8992, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.3101274136345724, | |
| "grad_norm": 0.9061586260795593, | |
| "learning_rate": 4.388888888888889e-05, | |
| "loss": 0.8966, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 3.362669118612899, | |
| "grad_norm": 0.7893356084823608, | |
| "learning_rate": 4.3703703703703705e-05, | |
| "loss": 0.8772, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.4152108235912255, | |
| "grad_norm": 1.1485052108764648, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 0.8935, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.4152108235912255, | |
| "eval_loss": 0.8316594362258911, | |
| "eval_runtime": 1.2559, | |
| "eval_samples_per_second": 7.962, | |
| "eval_steps_per_second": 3.981, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.467752528569552, | |
| "grad_norm": 1.2252397537231445, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.8877, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 3.5202942335478786, | |
| "grad_norm": 0.937276303768158, | |
| "learning_rate": 4.314814814814815e-05, | |
| "loss": 0.899, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 3.572835938526205, | |
| "grad_norm": 1.1383343935012817, | |
| "learning_rate": 4.296296296296296e-05, | |
| "loss": 0.884, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.6253776435045317, | |
| "grad_norm": 0.9624786376953125, | |
| "learning_rate": 4.277777777777778e-05, | |
| "loss": 0.8933, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 3.6779193484828583, | |
| "grad_norm": 1.2297359704971313, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.8902, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.6779193484828583, | |
| "eval_loss": 0.8357389569282532, | |
| "eval_runtime": 1.2564, | |
| "eval_samples_per_second": 7.959, | |
| "eval_steps_per_second": 3.979, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.730461053461185, | |
| "grad_norm": 1.3117122650146484, | |
| "learning_rate": 4.240740740740741e-05, | |
| "loss": 0.8833, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 3.7830027584395114, | |
| "grad_norm": 1.0405583381652832, | |
| "learning_rate": 4.222222222222222e-05, | |
| "loss": 0.8974, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.835544463417838, | |
| "grad_norm": 1.1837128400802612, | |
| "learning_rate": 4.203703703703704e-05, | |
| "loss": 0.9108, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 3.8880861683961645, | |
| "grad_norm": 1.070272445678711, | |
| "learning_rate": 4.185185185185185e-05, | |
| "loss": 0.888, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 3.940627873374491, | |
| "grad_norm": 1.2699558734893799, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.8996, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.940627873374491, | |
| "eval_loss": 0.8259842991828918, | |
| "eval_runtime": 1.2588, | |
| "eval_samples_per_second": 7.944, | |
| "eval_steps_per_second": 3.972, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.9931695783528176, | |
| "grad_norm": 1.1410374641418457, | |
| "learning_rate": 4.148148148148148e-05, | |
| "loss": 0.8931, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.045711283331144, | |
| "grad_norm": 0.8494368195533752, | |
| "learning_rate": 4.12962962962963e-05, | |
| "loss": 0.8735, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 4.09825298830947, | |
| "grad_norm": 1.6662884950637817, | |
| "learning_rate": 4.111111111111111e-05, | |
| "loss": 0.8706, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 4.150794693287797, | |
| "grad_norm": 1.103721022605896, | |
| "learning_rate": 4.092592592592593e-05, | |
| "loss": 0.8868, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 4.203336398266123, | |
| "grad_norm": 1.0457319021224976, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.8616, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.203336398266123, | |
| "eval_loss": 0.8296699523925781, | |
| "eval_runtime": 1.2594, | |
| "eval_samples_per_second": 7.941, | |
| "eval_steps_per_second": 3.97, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.25587810324445, | |
| "grad_norm": 1.038172721862793, | |
| "learning_rate": 4.055555555555556e-05, | |
| "loss": 0.8814, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 4.3084198082227765, | |
| "grad_norm": 1.0770879983901978, | |
| "learning_rate": 4.0370370370370374e-05, | |
| "loss": 0.8781, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 4.3609615132011035, | |
| "grad_norm": 0.9559742212295532, | |
| "learning_rate": 4.018518518518519e-05, | |
| "loss": 0.8631, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 4.41350321817943, | |
| "grad_norm": 1.2352747917175293, | |
| "learning_rate": 4e-05, | |
| "loss": 0.8619, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.466044923157757, | |
| "grad_norm": 1.0156300067901611, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 0.8589, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.466044923157757, | |
| "eval_loss": 0.8272676467895508, | |
| "eval_runtime": 1.239, | |
| "eval_samples_per_second": 8.071, | |
| "eval_steps_per_second": 4.035, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.518586628136083, | |
| "grad_norm": 1.3172954320907593, | |
| "learning_rate": 3.962962962962963e-05, | |
| "loss": 0.8782, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 4.57112833311441, | |
| "grad_norm": 1.0361146926879883, | |
| "learning_rate": 3.944444444444445e-05, | |
| "loss": 0.8929, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 4.623670038092736, | |
| "grad_norm": 1.0141929388046265, | |
| "learning_rate": 3.925925925925926e-05, | |
| "loss": 0.8701, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 4.676211743071063, | |
| "grad_norm": 1.4929876327514648, | |
| "learning_rate": 3.9074074074074076e-05, | |
| "loss": 0.8877, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 4.728753448049389, | |
| "grad_norm": 1.6893892288208008, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.8751, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.728753448049389, | |
| "eval_loss": 0.822452187538147, | |
| "eval_runtime": 1.246, | |
| "eval_samples_per_second": 8.026, | |
| "eval_steps_per_second": 4.013, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.781295153027716, | |
| "grad_norm": 0.993008017539978, | |
| "learning_rate": 3.8703703703703705e-05, | |
| "loss": 0.8689, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 4.833836858006042, | |
| "grad_norm": 0.8024290800094604, | |
| "learning_rate": 3.851851851851852e-05, | |
| "loss": 0.8945, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 4.886378562984369, | |
| "grad_norm": 0.8442456722259521, | |
| "learning_rate": 3.8333333333333334e-05, | |
| "loss": 0.8897, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 4.938920267962695, | |
| "grad_norm": 1.6505470275878906, | |
| "learning_rate": 3.814814814814815e-05, | |
| "loss": 0.8481, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 4.991461972941022, | |
| "grad_norm": 0.9656145572662354, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 0.8699, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.991461972941022, | |
| "eval_loss": 0.8228476643562317, | |
| "eval_runtime": 1.2383, | |
| "eval_samples_per_second": 8.075, | |
| "eval_steps_per_second": 4.038, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 5.044003677919348, | |
| "grad_norm": 1.1603432893753052, | |
| "learning_rate": 3.777777777777778e-05, | |
| "loss": 0.8731, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 5.096545382897675, | |
| "grad_norm": 0.8760092854499817, | |
| "learning_rate": 3.759259259259259e-05, | |
| "loss": 0.8268, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 5.149087087876001, | |
| "grad_norm": 1.50885009765625, | |
| "learning_rate": 3.740740740740741e-05, | |
| "loss": 0.8341, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 5.201628792854328, | |
| "grad_norm": 1.164678692817688, | |
| "learning_rate": 3.722222222222222e-05, | |
| "loss": 0.8511, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 5.254170497832654, | |
| "grad_norm": 1.2179008722305298, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.863, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.254170497832654, | |
| "eval_loss": 0.8227807283401489, | |
| "eval_runtime": 1.2397, | |
| "eval_samples_per_second": 8.067, | |
| "eval_steps_per_second": 4.033, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.306712202810981, | |
| "grad_norm": 1.1593024730682373, | |
| "learning_rate": 3.685185185185185e-05, | |
| "loss": 0.8481, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 5.3592539077893075, | |
| "grad_norm": 1.1822712421417236, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.8435, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 5.4117956127676345, | |
| "grad_norm": 1.1157796382904053, | |
| "learning_rate": 3.648148148148148e-05, | |
| "loss": 0.8265, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 5.464337317745961, | |
| "grad_norm": 1.2513401508331299, | |
| "learning_rate": 3.62962962962963e-05, | |
| "loss": 0.8916, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 5.516879022724288, | |
| "grad_norm": 1.07020902633667, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.855, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.516879022724288, | |
| "eval_loss": 0.8260787725448608, | |
| "eval_runtime": 1.2426, | |
| "eval_samples_per_second": 8.048, | |
| "eval_steps_per_second": 4.024, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.569420727702614, | |
| "grad_norm": 1.281821370124817, | |
| "learning_rate": 3.592592592592593e-05, | |
| "loss": 0.8375, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 5.621962432680941, | |
| "grad_norm": 1.279232144355774, | |
| "learning_rate": 3.574074074074074e-05, | |
| "loss": 0.8374, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 5.674504137659267, | |
| "grad_norm": 1.6183032989501953, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.8396, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 5.727045842637594, | |
| "grad_norm": 1.5569584369659424, | |
| "learning_rate": 3.537037037037037e-05, | |
| "loss": 0.8474, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 5.77958754761592, | |
| "grad_norm": 0.9284697771072388, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.8447, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.77958754761592, | |
| "eval_loss": 0.82489413022995, | |
| "eval_runtime": 1.2928, | |
| "eval_samples_per_second": 7.735, | |
| "eval_steps_per_second": 3.868, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.832129252594247, | |
| "grad_norm": 1.129302740097046, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.8425, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 5.884670957572573, | |
| "grad_norm": 1.1644898653030396, | |
| "learning_rate": 3.481481481481482e-05, | |
| "loss": 0.8545, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 5.9372126625509, | |
| "grad_norm": 1.5778745412826538, | |
| "learning_rate": 3.4629629629629626e-05, | |
| "loss": 0.8488, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 5.989754367529226, | |
| "grad_norm": 1.0982458591461182, | |
| "learning_rate": 3.444444444444445e-05, | |
| "loss": 0.8368, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 6.042296072507553, | |
| "grad_norm": 1.3936185836791992, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 0.8165, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.042296072507553, | |
| "eval_loss": 0.8229165077209473, | |
| "eval_runtime": 1.2812, | |
| "eval_samples_per_second": 7.805, | |
| "eval_steps_per_second": 3.903, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.094837777485879, | |
| "grad_norm": 0.8609622716903687, | |
| "learning_rate": 3.4074074074074077e-05, | |
| "loss": 0.813, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 6.147379482464206, | |
| "grad_norm": 1.75618314743042, | |
| "learning_rate": 3.388888888888889e-05, | |
| "loss": 0.8285, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 6.199921187442532, | |
| "grad_norm": 1.5755348205566406, | |
| "learning_rate": 3.3703703703703706e-05, | |
| "loss": 0.8061, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 6.252462892420859, | |
| "grad_norm": 1.1621313095092773, | |
| "learning_rate": 3.351851851851852e-05, | |
| "loss": 0.8349, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 6.3050045973991855, | |
| "grad_norm": 1.3047136068344116, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.8247, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.3050045973991855, | |
| "eval_loss": 0.8217476010322571, | |
| "eval_runtime": 1.249, | |
| "eval_samples_per_second": 8.007, | |
| "eval_steps_per_second": 4.003, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.3575463023775125, | |
| "grad_norm": 1.161156415939331, | |
| "learning_rate": 3.314814814814815e-05, | |
| "loss": 0.8248, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 6.410088007355839, | |
| "grad_norm": 1.095155954360962, | |
| "learning_rate": 3.2962962962962964e-05, | |
| "loss": 0.8275, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 6.462629712334166, | |
| "grad_norm": 1.3355261087417603, | |
| "learning_rate": 3.277777777777778e-05, | |
| "loss": 0.833, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 6.515171417312492, | |
| "grad_norm": 1.5305536985397339, | |
| "learning_rate": 3.25925925925926e-05, | |
| "loss": 0.8259, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 6.567713122290819, | |
| "grad_norm": 2.5994789600372314, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.8459, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.567713122290819, | |
| "eval_loss": 0.8286579251289368, | |
| "eval_runtime": 1.2728, | |
| "eval_samples_per_second": 7.856, | |
| "eval_steps_per_second": 3.928, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.620254827269145, | |
| "grad_norm": 1.442550539970398, | |
| "learning_rate": 3.222222222222223e-05, | |
| "loss": 0.8294, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 6.672796532247471, | |
| "grad_norm": 1.1451183557510376, | |
| "learning_rate": 3.203703703703704e-05, | |
| "loss": 0.8284, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 6.725338237225798, | |
| "grad_norm": 1.0634784698486328, | |
| "learning_rate": 3.185185185185185e-05, | |
| "loss": 0.823, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 6.777879942204125, | |
| "grad_norm": 1.2910544872283936, | |
| "learning_rate": 3.1666666666666666e-05, | |
| "loss": 0.8248, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 6.830421647182451, | |
| "grad_norm": 1.1855148077011108, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.854, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.830421647182451, | |
| "eval_loss": 0.8177132606506348, | |
| "eval_runtime": 1.2889, | |
| "eval_samples_per_second": 7.759, | |
| "eval_steps_per_second": 3.879, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.882963352160777, | |
| "grad_norm": 1.2636213302612305, | |
| "learning_rate": 3.1296296296296295e-05, | |
| "loss": 0.8429, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 6.935505057139104, | |
| "grad_norm": 1.2894644737243652, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.8279, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 6.988046762117431, | |
| "grad_norm": 1.64339280128479, | |
| "learning_rate": 3.0925925925925924e-05, | |
| "loss": 0.8217, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 7.040588467095757, | |
| "grad_norm": 0.9605098962783813, | |
| "learning_rate": 3.074074074074074e-05, | |
| "loss": 0.8375, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 7.093130172074083, | |
| "grad_norm": 1.4828298091888428, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.8079, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.093130172074083, | |
| "eval_loss": 0.8257712125778198, | |
| "eval_runtime": 1.2505, | |
| "eval_samples_per_second": 7.997, | |
| "eval_steps_per_second": 3.998, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.14567187705241, | |
| "grad_norm": 1.3807919025421143, | |
| "learning_rate": 3.037037037037037e-05, | |
| "loss": 0.8096, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 7.1982135820307365, | |
| "grad_norm": 1.5678062438964844, | |
| "learning_rate": 3.018518518518519e-05, | |
| "loss": 0.7869, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 7.2507552870090635, | |
| "grad_norm": 1.0710649490356445, | |
| "learning_rate": 3e-05, | |
| "loss": 0.813, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 7.30329699198739, | |
| "grad_norm": 1.150152564048767, | |
| "learning_rate": 2.981481481481482e-05, | |
| "loss": 0.7974, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 7.355838696965717, | |
| "grad_norm": 1.4336960315704346, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.8126, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.355838696965717, | |
| "eval_loss": 0.8250346183776855, | |
| "eval_runtime": 1.2966, | |
| "eval_samples_per_second": 7.712, | |
| "eval_steps_per_second": 3.856, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.408380401944043, | |
| "grad_norm": 1.454750895500183, | |
| "learning_rate": 2.9444444444444448e-05, | |
| "loss": 0.8534, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 7.46092210692237, | |
| "grad_norm": 1.3246815204620361, | |
| "learning_rate": 2.925925925925926e-05, | |
| "loss": 0.8313, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 7.513463811900696, | |
| "grad_norm": 1.116432547569275, | |
| "learning_rate": 2.9074074074074077e-05, | |
| "loss": 0.8085, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 7.566005516879023, | |
| "grad_norm": 1.0311377048492432, | |
| "learning_rate": 2.8888888888888888e-05, | |
| "loss": 0.7984, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 7.618547221857349, | |
| "grad_norm": 1.5071169137954712, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.817, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.618547221857349, | |
| "eval_loss": 0.8244579434394836, | |
| "eval_runtime": 1.2449, | |
| "eval_samples_per_second": 8.032, | |
| "eval_steps_per_second": 4.016, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.671088926835676, | |
| "grad_norm": 1.2359856367111206, | |
| "learning_rate": 2.851851851851852e-05, | |
| "loss": 0.807, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 7.723630631814002, | |
| "grad_norm": 1.2495603561401367, | |
| "learning_rate": 2.8333333333333335e-05, | |
| "loss": 0.8035, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 7.776172336792329, | |
| "grad_norm": 1.0622730255126953, | |
| "learning_rate": 2.814814814814815e-05, | |
| "loss": 0.8199, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 7.828714041770655, | |
| "grad_norm": 1.097486138343811, | |
| "learning_rate": 2.7962962962962965e-05, | |
| "loss": 0.8107, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 7.881255746748982, | |
| "grad_norm": 1.1956055164337158, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.7962, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.881255746748982, | |
| "eval_loss": 0.8224829435348511, | |
| "eval_runtime": 1.2586, | |
| "eval_samples_per_second": 7.946, | |
| "eval_steps_per_second": 3.973, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.933797451727308, | |
| "grad_norm": 1.154567837715149, | |
| "learning_rate": 2.7592592592592594e-05, | |
| "loss": 0.8278, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 7.986339156705635, | |
| "grad_norm": 1.4666606187820435, | |
| "learning_rate": 2.7407407407407408e-05, | |
| "loss": 0.8157, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 8.038880861683962, | |
| "grad_norm": 1.115820050239563, | |
| "learning_rate": 2.7222222222222223e-05, | |
| "loss": 0.7899, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 8.091422566662288, | |
| "grad_norm": 1.4497051239013672, | |
| "learning_rate": 2.7037037037037037e-05, | |
| "loss": 0.7809, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 8.143964271640614, | |
| "grad_norm": 1.9520905017852783, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.7972, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 8.143964271640614, | |
| "eval_loss": 0.8295345306396484, | |
| "eval_runtime": 1.2468, | |
| "eval_samples_per_second": 8.02, | |
| "eval_steps_per_second": 4.01, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 8.19650597661894, | |
| "grad_norm": 1.1650198698043823, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.7869, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 8.249047681597268, | |
| "grad_norm": 0.9997352957725525, | |
| "learning_rate": 2.6481481481481485e-05, | |
| "loss": 0.7786, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 8.301589386575595, | |
| "grad_norm": 1.173376202583313, | |
| "learning_rate": 2.6296296296296296e-05, | |
| "loss": 0.7903, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 8.35413109155392, | |
| "grad_norm": 1.3240666389465332, | |
| "learning_rate": 2.6111111111111114e-05, | |
| "loss": 0.7874, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 8.406672796532247, | |
| "grad_norm": 1.077017903327942, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.7835, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.406672796532247, | |
| "eval_loss": 0.8272402882575989, | |
| "eval_runtime": 1.2976, | |
| "eval_samples_per_second": 7.706, | |
| "eval_steps_per_second": 3.853, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.459214501510575, | |
| "grad_norm": 1.0706350803375244, | |
| "learning_rate": 2.5740740740740743e-05, | |
| "loss": 0.8083, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 8.5117562064889, | |
| "grad_norm": 1.0505809783935547, | |
| "learning_rate": 2.5555555555555554e-05, | |
| "loss": 0.8085, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 8.564297911467227, | |
| "grad_norm": 1.0813027620315552, | |
| "learning_rate": 2.5370370370370372e-05, | |
| "loss": 0.7943, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 8.616839616445553, | |
| "grad_norm": 1.4488904476165771, | |
| "learning_rate": 2.5185185185185183e-05, | |
| "loss": 0.7885, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 8.66938132142388, | |
| "grad_norm": 1.1785513162612915, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7867, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.66938132142388, | |
| "eval_loss": 0.8290830850601196, | |
| "eval_runtime": 1.2989, | |
| "eval_samples_per_second": 7.699, | |
| "eval_steps_per_second": 3.849, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.721923026402207, | |
| "grad_norm": 1.331755518913269, | |
| "learning_rate": 2.4814814814814816e-05, | |
| "loss": 0.7887, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 8.774464731380533, | |
| "grad_norm": 1.0079772472381592, | |
| "learning_rate": 2.462962962962963e-05, | |
| "loss": 0.7831, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 8.82700643635886, | |
| "grad_norm": 1.6257929801940918, | |
| "learning_rate": 2.4444444444444445e-05, | |
| "loss": 0.788, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 8.879548141337187, | |
| "grad_norm": 1.5751090049743652, | |
| "learning_rate": 2.425925925925926e-05, | |
| "loss": 0.7864, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 8.932089846315513, | |
| "grad_norm": 1.0488544702529907, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.8083, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.932089846315513, | |
| "eval_loss": 0.8188499212265015, | |
| "eval_runtime": 1.2935, | |
| "eval_samples_per_second": 7.731, | |
| "eval_steps_per_second": 3.865, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.98463155129384, | |
| "grad_norm": 1.8487260341644287, | |
| "learning_rate": 2.3888888888888892e-05, | |
| "loss": 0.7956, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 9.037173256272165, | |
| "grad_norm": 1.395914912223816, | |
| "learning_rate": 2.3703703703703707e-05, | |
| "loss": 0.7921, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 9.089714961250493, | |
| "grad_norm": 1.341793417930603, | |
| "learning_rate": 2.351851851851852e-05, | |
| "loss": 0.7975, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 9.14225666622882, | |
| "grad_norm": 1.0076344013214111, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.773, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 9.194798371207145, | |
| "grad_norm": 1.081777811050415, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.7654, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 9.194798371207145, | |
| "eval_loss": 0.8274938464164734, | |
| "eval_runtime": 1.2836, | |
| "eval_samples_per_second": 7.79, | |
| "eval_steps_per_second": 3.895, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 9.247340076185472, | |
| "grad_norm": 1.1430171728134155, | |
| "learning_rate": 2.2962962962962965e-05, | |
| "loss": 0.7956, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 9.2998817811638, | |
| "grad_norm": 1.0450961589813232, | |
| "learning_rate": 2.277777777777778e-05, | |
| "loss": 0.7774, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 9.352423486142126, | |
| "grad_norm": 1.0086592435836792, | |
| "learning_rate": 2.2592592592592594e-05, | |
| "loss": 0.7739, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 9.404965191120452, | |
| "grad_norm": 1.7970887422561646, | |
| "learning_rate": 2.240740740740741e-05, | |
| "loss": 0.7793, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 9.457506896098778, | |
| "grad_norm": 1.5623388290405273, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.7793, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.457506896098778, | |
| "eval_loss": 0.8288339376449585, | |
| "eval_runtime": 1.2843, | |
| "eval_samples_per_second": 7.786, | |
| "eval_steps_per_second": 3.893, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.510048601077106, | |
| "grad_norm": 1.1271198987960815, | |
| "learning_rate": 2.2037037037037038e-05, | |
| "loss": 0.7693, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 9.562590306055432, | |
| "grad_norm": 1.0703665018081665, | |
| "learning_rate": 2.1851851851851852e-05, | |
| "loss": 0.7734, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 9.615132011033758, | |
| "grad_norm": 1.5558685064315796, | |
| "learning_rate": 2.1666666666666667e-05, | |
| "loss": 0.7778, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 9.667673716012084, | |
| "grad_norm": 1.932632565498352, | |
| "learning_rate": 2.148148148148148e-05, | |
| "loss": 0.7842, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 9.720215420990412, | |
| "grad_norm": 2.2928659915924072, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.7761, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.720215420990412, | |
| "eval_loss": 0.8304858207702637, | |
| "eval_runtime": 1.2431, | |
| "eval_samples_per_second": 8.045, | |
| "eval_steps_per_second": 4.022, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.772757125968738, | |
| "grad_norm": 1.588517189025879, | |
| "learning_rate": 2.111111111111111e-05, | |
| "loss": 0.7738, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 9.825298830947064, | |
| "grad_norm": 1.4118093252182007, | |
| "learning_rate": 2.0925925925925925e-05, | |
| "loss": 0.7852, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 9.87784053592539, | |
| "grad_norm": 1.7786134481430054, | |
| "learning_rate": 2.074074074074074e-05, | |
| "loss": 0.789, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 9.930382240903718, | |
| "grad_norm": 1.4427546262741089, | |
| "learning_rate": 2.0555555555555555e-05, | |
| "loss": 0.8028, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 9.982923945882044, | |
| "grad_norm": 1.2574381828308105, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.7694, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 9.982923945882044, | |
| "eval_loss": 0.8281936645507812, | |
| "eval_runtime": 1.2455, | |
| "eval_samples_per_second": 8.029, | |
| "eval_steps_per_second": 4.015, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 10.03546565086037, | |
| "grad_norm": 1.5743870735168457, | |
| "learning_rate": 2.0185185185185187e-05, | |
| "loss": 0.7619, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 10.088007355838696, | |
| "grad_norm": 1.350224494934082, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7596, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 10.140549060817024, | |
| "grad_norm": 1.1520729064941406, | |
| "learning_rate": 1.9814814814814816e-05, | |
| "loss": 0.7561, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 10.19309076579535, | |
| "grad_norm": 1.2705488204956055, | |
| "learning_rate": 1.962962962962963e-05, | |
| "loss": 0.7659, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 10.245632470773677, | |
| "grad_norm": 1.1752071380615234, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.7597, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 10.245632470773677, | |
| "eval_loss": 0.8358650207519531, | |
| "eval_runtime": 1.2389, | |
| "eval_samples_per_second": 8.072, | |
| "eval_steps_per_second": 4.036, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 10.298174175752003, | |
| "grad_norm": 1.3173915147781372, | |
| "learning_rate": 1.925925925925926e-05, | |
| "loss": 0.7787, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 10.35071588073033, | |
| "grad_norm": 1.3181281089782715, | |
| "learning_rate": 1.9074074074074075e-05, | |
| "loss": 0.7559, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 10.403257585708657, | |
| "grad_norm": 1.1058646440505981, | |
| "learning_rate": 1.888888888888889e-05, | |
| "loss": 0.7659, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 10.455799290686983, | |
| "grad_norm": 1.2898966073989868, | |
| "learning_rate": 1.8703703703703704e-05, | |
| "loss": 0.7642, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 10.508340995665309, | |
| "grad_norm": 1.7255585193634033, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.7659, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.508340995665309, | |
| "eval_loss": 0.8337475061416626, | |
| "eval_runtime": 1.2724, | |
| "eval_samples_per_second": 7.859, | |
| "eval_steps_per_second": 3.93, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.560882700643635, | |
| "grad_norm": 1.101854681968689, | |
| "learning_rate": 1.8333333333333333e-05, | |
| "loss": 0.7754, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 10.613424405621963, | |
| "grad_norm": 1.6675972938537598, | |
| "learning_rate": 1.814814814814815e-05, | |
| "loss": 0.7597, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 10.665966110600289, | |
| "grad_norm": 1.428888201713562, | |
| "learning_rate": 1.7962962962962965e-05, | |
| "loss": 0.7714, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 10.718507815578615, | |
| "grad_norm": 1.2556451559066772, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.7628, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 10.771049520556943, | |
| "grad_norm": 1.817168951034546, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.7652, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 10.771049520556943, | |
| "eval_loss": 0.8289246559143066, | |
| "eval_runtime": 1.2657, | |
| "eval_samples_per_second": 7.901, | |
| "eval_steps_per_second": 3.95, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 10.823591225535269, | |
| "grad_norm": 0.9452287554740906, | |
| "learning_rate": 1.740740740740741e-05, | |
| "loss": 0.7661, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 10.876132930513595, | |
| "grad_norm": 1.0433977842330933, | |
| "learning_rate": 1.7222222222222224e-05, | |
| "loss": 0.7662, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 10.928674635491921, | |
| "grad_norm": 1.7316306829452515, | |
| "learning_rate": 1.7037037037037038e-05, | |
| "loss": 0.7694, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 10.981216340470247, | |
| "grad_norm": 1.1855412721633911, | |
| "learning_rate": 1.6851851851851853e-05, | |
| "loss": 0.7608, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 11.033758045448575, | |
| "grad_norm": 1.2221699953079224, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.7538, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 11.033758045448575, | |
| "eval_loss": 0.8385075330734253, | |
| "eval_runtime": 1.2472, | |
| "eval_samples_per_second": 8.018, | |
| "eval_steps_per_second": 4.009, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 11.086299750426901, | |
| "grad_norm": 1.1642967462539673, | |
| "learning_rate": 1.6481481481481482e-05, | |
| "loss": 0.7348, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 11.138841455405228, | |
| "grad_norm": 1.7363336086273193, | |
| "learning_rate": 1.62962962962963e-05, | |
| "loss": 0.756, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 11.191383160383554, | |
| "grad_norm": 1.3126277923583984, | |
| "learning_rate": 1.6111111111111115e-05, | |
| "loss": 0.7515, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 11.243924865361882, | |
| "grad_norm": 2.1372787952423096, | |
| "learning_rate": 1.5925925925925926e-05, | |
| "loss": 0.7487, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 11.296466570340208, | |
| "grad_norm": 1.5635650157928467, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.7554, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 11.296466570340208, | |
| "eval_loss": 0.8373192548751831, | |
| "eval_runtime": 1.2681, | |
| "eval_samples_per_second": 7.886, | |
| "eval_steps_per_second": 3.943, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 11.349008275318534, | |
| "grad_norm": 1.708914875984192, | |
| "learning_rate": 1.5555555555555555e-05, | |
| "loss": 0.7648, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 11.40154998029686, | |
| "grad_norm": 1.1601842641830444, | |
| "learning_rate": 1.537037037037037e-05, | |
| "loss": 0.7614, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 11.454091685275188, | |
| "grad_norm": 1.1674526929855347, | |
| "learning_rate": 1.5185185185185186e-05, | |
| "loss": 0.7653, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 11.506633390253514, | |
| "grad_norm": 1.2474111318588257, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.7596, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 11.55917509523184, | |
| "grad_norm": 1.2556248903274536, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.7445, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.55917509523184, | |
| "eval_loss": 0.8322497606277466, | |
| "eval_runtime": 1.2428, | |
| "eval_samples_per_second": 8.046, | |
| "eval_steps_per_second": 4.023, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.611716800210166, | |
| "grad_norm": 1.5506947040557861, | |
| "learning_rate": 1.462962962962963e-05, | |
| "loss": 0.7449, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 11.664258505188494, | |
| "grad_norm": 1.124089241027832, | |
| "learning_rate": 1.4444444444444444e-05, | |
| "loss": 0.7613, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 11.71680021016682, | |
| "grad_norm": 1.5663940906524658, | |
| "learning_rate": 1.425925925925926e-05, | |
| "loss": 0.7447, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 11.769341915145146, | |
| "grad_norm": 1.4085638523101807, | |
| "learning_rate": 1.4074074074074075e-05, | |
| "loss": 0.745, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 11.821883620123472, | |
| "grad_norm": 1.8693653345108032, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.7615, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 11.821883620123472, | |
| "eval_loss": 0.8351184129714966, | |
| "eval_runtime": 1.2433, | |
| "eval_samples_per_second": 8.043, | |
| "eval_steps_per_second": 4.022, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 11.8744253251018, | |
| "grad_norm": 1.807320237159729, | |
| "learning_rate": 1.3703703703703704e-05, | |
| "loss": 0.7605, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 11.926967030080126, | |
| "grad_norm": 1.1300692558288574, | |
| "learning_rate": 1.3518518518518519e-05, | |
| "loss": 0.7604, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 11.979508735058452, | |
| "grad_norm": 1.5714143514633179, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.7462, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 12.032050440036778, | |
| "grad_norm": 1.1730937957763672, | |
| "learning_rate": 1.3148148148148148e-05, | |
| "loss": 0.7655, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 12.084592145015106, | |
| "grad_norm": 1.1946797370910645, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.7326, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 12.084592145015106, | |
| "eval_loss": 0.8375317454338074, | |
| "eval_runtime": 1.2867, | |
| "eval_samples_per_second": 7.772, | |
| "eval_steps_per_second": 3.886, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 12.137133849993432, | |
| "grad_norm": 1.1328163146972656, | |
| "learning_rate": 1.2777777777777777e-05, | |
| "loss": 0.7439, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 12.189675554971759, | |
| "grad_norm": 1.026856780052185, | |
| "learning_rate": 1.2592592592592592e-05, | |
| "loss": 0.7589, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 12.242217259950085, | |
| "grad_norm": 1.920116662979126, | |
| "learning_rate": 1.2407407407407408e-05, | |
| "loss": 0.7458, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 12.294758964928413, | |
| "grad_norm": 1.3238016366958618, | |
| "learning_rate": 1.2222222222222222e-05, | |
| "loss": 0.74, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 12.347300669906739, | |
| "grad_norm": 1.192124843597412, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.7369, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 12.347300669906739, | |
| "eval_loss": 0.8405577540397644, | |
| "eval_runtime": 1.2401, | |
| "eval_samples_per_second": 8.064, | |
| "eval_steps_per_second": 4.032, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 12.399842374885065, | |
| "grad_norm": 1.3931485414505005, | |
| "learning_rate": 1.1851851851851853e-05, | |
| "loss": 0.7619, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 12.452384079863391, | |
| "grad_norm": 1.071673035621643, | |
| "learning_rate": 1.1666666666666668e-05, | |
| "loss": 0.753, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 12.504925784841719, | |
| "grad_norm": 1.2407554388046265, | |
| "learning_rate": 1.1481481481481482e-05, | |
| "loss": 0.7382, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 12.557467489820045, | |
| "grad_norm": 1.2360953092575073, | |
| "learning_rate": 1.1296296296296297e-05, | |
| "loss": 0.7348, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 12.610009194798371, | |
| "grad_norm": 1.2687872648239136, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.75, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 12.610009194798371, | |
| "eval_loss": 0.8404332399368286, | |
| "eval_runtime": 1.2513, | |
| "eval_samples_per_second": 7.992, | |
| "eval_steps_per_second": 3.996, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 12.662550899776697, | |
| "grad_norm": 1.1866815090179443, | |
| "learning_rate": 1.0925925925925926e-05, | |
| "loss": 0.7419, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 12.715092604755025, | |
| "grad_norm": 1.715956687927246, | |
| "learning_rate": 1.074074074074074e-05, | |
| "loss": 0.7406, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 12.767634309733351, | |
| "grad_norm": 1.1529712677001953, | |
| "learning_rate": 1.0555555555555555e-05, | |
| "loss": 0.7516, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 12.820176014711677, | |
| "grad_norm": 1.2465269565582275, | |
| "learning_rate": 1.037037037037037e-05, | |
| "loss": 0.7407, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 12.872717719690003, | |
| "grad_norm": 1.1018378734588623, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.7514, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 12.872717719690003, | |
| "eval_loss": 0.8341609835624695, | |
| "eval_runtime": 1.2427, | |
| "eval_samples_per_second": 8.047, | |
| "eval_steps_per_second": 4.024, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 12.925259424668331, | |
| "grad_norm": 1.2097246646881104, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7298, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 12.977801129646657, | |
| "grad_norm": 1.5760364532470703, | |
| "learning_rate": 9.814814814814815e-06, | |
| "loss": 0.7538, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 13.030342834624983, | |
| "grad_norm": 1.2670038938522339, | |
| "learning_rate": 9.62962962962963e-06, | |
| "loss": 0.7437, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 13.08288453960331, | |
| "grad_norm": 1.178148627281189, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.7392, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 13.135426244581637, | |
| "grad_norm": 1.2803364992141724, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.7178, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 13.135426244581637, | |
| "eval_loss": 0.8406384587287903, | |
| "eval_runtime": 1.2436, | |
| "eval_samples_per_second": 8.041, | |
| "eval_steps_per_second": 4.02, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 13.187967949559964, | |
| "grad_norm": 1.3507287502288818, | |
| "learning_rate": 9.074074074074075e-06, | |
| "loss": 0.7473, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 13.24050965453829, | |
| "grad_norm": 1.1901633739471436, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 0.7386, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 13.293051359516616, | |
| "grad_norm": 1.25368332862854, | |
| "learning_rate": 8.703703703703705e-06, | |
| "loss": 0.7426, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 13.345593064494944, | |
| "grad_norm": 1.4581819772720337, | |
| "learning_rate": 8.518518518518519e-06, | |
| "loss": 0.7479, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 13.39813476947327, | |
| "grad_norm": 1.201744556427002, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.7266, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 13.39813476947327, | |
| "eval_loss": 0.8411173820495605, | |
| "eval_runtime": 1.2635, | |
| "eval_samples_per_second": 7.915, | |
| "eval_steps_per_second": 3.957, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 13.450676474451596, | |
| "grad_norm": 1.3724862337112427, | |
| "learning_rate": 8.14814814814815e-06, | |
| "loss": 0.7392, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 13.503218179429922, | |
| "grad_norm": 1.2667688131332397, | |
| "learning_rate": 7.962962962962963e-06, | |
| "loss": 0.731, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 13.55575988440825, | |
| "grad_norm": 1.3549320697784424, | |
| "learning_rate": 7.777777777777777e-06, | |
| "loss": 0.7307, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 13.608301589386576, | |
| "grad_norm": 1.3319281339645386, | |
| "learning_rate": 7.592592592592593e-06, | |
| "loss": 0.7315, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 13.660843294364902, | |
| "grad_norm": 1.3759852647781372, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.7338, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 13.660843294364902, | |
| "eval_loss": 0.8397306203842163, | |
| "eval_runtime": 1.2452, | |
| "eval_samples_per_second": 8.031, | |
| "eval_steps_per_second": 4.015, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 13.713384999343228, | |
| "grad_norm": 1.452658772468567, | |
| "learning_rate": 7.222222222222222e-06, | |
| "loss": 0.7339, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 13.765926704321554, | |
| "grad_norm": 1.1617493629455566, | |
| "learning_rate": 7.0370370370370375e-06, | |
| "loss": 0.737, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 13.818468409299882, | |
| "grad_norm": 1.628431797027588, | |
| "learning_rate": 6.851851851851852e-06, | |
| "loss": 0.7437, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 13.871010114278208, | |
| "grad_norm": 1.4544808864593506, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.7411, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 13.923551819256534, | |
| "grad_norm": 1.7027857303619385, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.7432, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 13.923551819256534, | |
| "eval_loss": 0.8381926417350769, | |
| "eval_runtime": 1.2458, | |
| "eval_samples_per_second": 8.027, | |
| "eval_steps_per_second": 4.013, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 13.976093524234862, | |
| "grad_norm": 1.320162296295166, | |
| "learning_rate": 6.296296296296296e-06, | |
| "loss": 0.7463, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 14.028635229213188, | |
| "grad_norm": 1.2044928073883057, | |
| "learning_rate": 6.111111111111111e-06, | |
| "loss": 0.7358, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 14.081176934191515, | |
| "grad_norm": 1.0798685550689697, | |
| "learning_rate": 5.925925925925927e-06, | |
| "loss": 0.7416, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 14.13371863916984, | |
| "grad_norm": 1.4219785928726196, | |
| "learning_rate": 5.740740740740741e-06, | |
| "loss": 0.7256, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 14.186260344148167, | |
| "grad_norm": 1.1347541809082031, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.7322, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 14.186260344148167, | |
| "eval_loss": 0.8432482481002808, | |
| "eval_runtime": 1.2881, | |
| "eval_samples_per_second": 7.763, | |
| "eval_steps_per_second": 3.882, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 14.238802049126495, | |
| "grad_norm": 1.5265473127365112, | |
| "learning_rate": 5.37037037037037e-06, | |
| "loss": 0.7281, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 14.29134375410482, | |
| "grad_norm": 1.0720326900482178, | |
| "learning_rate": 5.185185185185185e-06, | |
| "loss": 0.731, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 14.343885459083147, | |
| "grad_norm": 1.5150460004806519, | |
| "learning_rate": 5e-06, | |
| "loss": 0.719, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 14.396427164061473, | |
| "grad_norm": 1.2847768068313599, | |
| "learning_rate": 4.814814814814815e-06, | |
| "loss": 0.7307, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 14.4489688690398, | |
| "grad_norm": 1.4064009189605713, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.7293, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 14.4489688690398, | |
| "eval_loss": 0.8425260782241821, | |
| "eval_runtime": 1.247, | |
| "eval_samples_per_second": 8.019, | |
| "eval_steps_per_second": 4.01, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 14.501510574018127, | |
| "grad_norm": 1.1766741275787354, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 0.7406, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 14.554052278996453, | |
| "grad_norm": 1.3252798318862915, | |
| "learning_rate": 4.2592592592592596e-06, | |
| "loss": 0.7377, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 14.60659398397478, | |
| "grad_norm": 1.2687978744506836, | |
| "learning_rate": 4.074074074074075e-06, | |
| "loss": 0.7303, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 14.659135688953107, | |
| "grad_norm": 1.29690420627594, | |
| "learning_rate": 3.888888888888889e-06, | |
| "loss": 0.7349, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 14.711677393931433, | |
| "grad_norm": 1.4593523740768433, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.7327, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 14.711677393931433, | |
| "eval_loss": 0.842442512512207, | |
| "eval_runtime": 1.2608, | |
| "eval_samples_per_second": 7.931, | |
| "eval_steps_per_second": 3.966, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 14.76421909890976, | |
| "grad_norm": 1.1667600870132446, | |
| "learning_rate": 3.5185185185185187e-06, | |
| "loss": 0.7276, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 14.816760803888085, | |
| "grad_norm": 1.0612125396728516, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.7468, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 14.869302508866413, | |
| "grad_norm": 1.117440938949585, | |
| "learning_rate": 3.148148148148148e-06, | |
| "loss": 0.7189, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 14.92184421384474, | |
| "grad_norm": 1.1088405847549438, | |
| "learning_rate": 2.9629629629629633e-06, | |
| "loss": 0.732, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 14.974385918823065, | |
| "grad_norm": 1.381748914718628, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.7368, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 14.974385918823065, | |
| "eval_loss": 0.844028651714325, | |
| "eval_runtime": 1.2451, | |
| "eval_samples_per_second": 8.031, | |
| "eval_steps_per_second": 4.016, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 15.026927623801392, | |
| "grad_norm": 1.2738327980041504, | |
| "learning_rate": 2.5925925925925925e-06, | |
| "loss": 0.7411, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 15.07946932877972, | |
| "grad_norm": 1.450596809387207, | |
| "learning_rate": 2.4074074074074075e-06, | |
| "loss": 0.721, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 15.132011033758046, | |
| "grad_norm": 1.2589495182037354, | |
| "learning_rate": 2.2222222222222225e-06, | |
| "loss": 0.7134, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 15.184552738736372, | |
| "grad_norm": 1.1227418184280396, | |
| "learning_rate": 2.0370370370370375e-06, | |
| "loss": 0.7361, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 15.237094443714698, | |
| "grad_norm": 1.1688610315322876, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.722, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 15.237094443714698, | |
| "eval_loss": 0.8448740243911743, | |
| "eval_runtime": 1.2712, | |
| "eval_samples_per_second": 7.867, | |
| "eval_steps_per_second": 3.933, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 15.289636148693026, | |
| "grad_norm": 1.4582873582839966, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.7236, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 15.342177853671352, | |
| "grad_norm": 1.5905903577804565, | |
| "learning_rate": 1.4814814814814817e-06, | |
| "loss": 0.7332, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 15.394719558649678, | |
| "grad_norm": 2.0454487800598145, | |
| "learning_rate": 1.2962962962962962e-06, | |
| "loss": 0.7222, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 15.447261263628004, | |
| "grad_norm": 1.3177134990692139, | |
| "learning_rate": 1.1111111111111112e-06, | |
| "loss": 0.7362, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 15.499802968606332, | |
| "grad_norm": 1.0010744333267212, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.7304, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 15.499802968606332, | |
| "eval_loss": 0.84196537733078, | |
| "eval_runtime": 1.2445, | |
| "eval_samples_per_second": 8.035, | |
| "eval_steps_per_second": 4.018, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 15.552344673584658, | |
| "grad_norm": 1.296751618385315, | |
| "learning_rate": 7.407407407407408e-07, | |
| "loss": 0.719, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 15.604886378562984, | |
| "grad_norm": 1.198317050933838, | |
| "learning_rate": 5.555555555555556e-07, | |
| "loss": 0.7359, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 15.65742808354131, | |
| "grad_norm": 1.3005775213241577, | |
| "learning_rate": 3.703703703703704e-07, | |
| "loss": 0.7268, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 15.709969788519638, | |
| "grad_norm": 1.1755659580230713, | |
| "learning_rate": 1.851851851851852e-07, | |
| "loss": 0.7266, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 15.762511493497964, | |
| "grad_norm": 1.1558780670166016, | |
| "learning_rate": 0.0, | |
| "loss": 0.7389, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 15.762511493497964, | |
| "eval_loss": 0.8437487483024597, | |
| "eval_runtime": 1.263, | |
| "eval_samples_per_second": 7.918, | |
| "eval_steps_per_second": 3.959, | |
| "step": 30000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 30000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 16, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.658737511387988e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |