| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.835446953294154, | |
| "eval_steps": 200, | |
| "global_step": 12400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.6350187063217163, | |
| "learning_rate": 4.977888397519945e-05, | |
| "loss": 1.2168, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 1.0502910614013672, | |
| "eval_runtime": 524.206, | |
| "eval_samples_per_second": 12.058, | |
| "eval_steps_per_second": 1.509, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 1.5651005506515503, | |
| "learning_rate": 4.955665681459588e-05, | |
| "loss": 0.8535, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.9054814577102661, | |
| "eval_runtime": 523.7675, | |
| "eval_samples_per_second": 12.068, | |
| "eval_steps_per_second": 1.51, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.4049625396728516, | |
| "learning_rate": 4.9334429653992315e-05, | |
| "loss": 0.7543, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.8259416818618774, | |
| "eval_runtime": 524.2403, | |
| "eval_samples_per_second": 12.057, | |
| "eval_steps_per_second": 1.509, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.6197569370269775, | |
| "learning_rate": 4.911220249338874e-05, | |
| "loss": 0.7007, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.7792428135871887, | |
| "eval_runtime": 523.8718, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 1.51, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.3332139253616333, | |
| "learning_rate": 4.888997533278518e-05, | |
| "loss": 0.653, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.7383475303649902, | |
| "eval_runtime": 524.1649, | |
| "eval_samples_per_second": 12.059, | |
| "eval_steps_per_second": 1.509, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.3490664958953857, | |
| "learning_rate": 4.8667748172181604e-05, | |
| "loss": 0.6267, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.7095275521278381, | |
| "eval_runtime": 523.9699, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 1.2756898403167725, | |
| "learning_rate": 4.844552101157804e-05, | |
| "loss": 0.5969, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 0.6869618892669678, | |
| "eval_runtime": 524.0331, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.2612615823745728, | |
| "learning_rate": 4.8223293850974466e-05, | |
| "loss": 0.5771, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 0.6662651300430298, | |
| "eval_runtime": 524.2978, | |
| "eval_samples_per_second": 12.056, | |
| "eval_steps_per_second": 1.509, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.1881535053253174, | |
| "learning_rate": 4.80010666903709e-05, | |
| "loss": 0.5632, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.6456959247589111, | |
| "eval_runtime": 524.0526, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.1880247592926025, | |
| "learning_rate": 4.7778839529767335e-05, | |
| "loss": 0.5485, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_loss": 0.6321355700492859, | |
| "eval_runtime": 524.3064, | |
| "eval_samples_per_second": 12.056, | |
| "eval_steps_per_second": 1.509, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 1.1281764507293701, | |
| "learning_rate": 4.755661236916376e-05, | |
| "loss": 0.5412, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.6182820796966553, | |
| "eval_runtime": 524.0853, | |
| "eval_samples_per_second": 12.061, | |
| "eval_steps_per_second": 1.509, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.2166435718536377, | |
| "learning_rate": 4.73343852085602e-05, | |
| "loss": 0.5244, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_loss": 0.6053413152694702, | |
| "eval_runtime": 524.1727, | |
| "eval_samples_per_second": 12.059, | |
| "eval_steps_per_second": 1.509, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.0197983980178833, | |
| "learning_rate": 4.7112158047956625e-05, | |
| "loss": 0.5161, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.5936387181282043, | |
| "eval_runtime": 524.5444, | |
| "eval_samples_per_second": 12.05, | |
| "eval_steps_per_second": 1.508, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 1.0752451419830322, | |
| "learning_rate": 4.688993088735306e-05, | |
| "loss": 0.5043, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 0.583480715751648, | |
| "eval_runtime": 524.1844, | |
| "eval_samples_per_second": 12.059, | |
| "eval_steps_per_second": 1.509, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.186328649520874, | |
| "learning_rate": 4.666770372674949e-05, | |
| "loss": 0.495, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 0.5726756453514099, | |
| "eval_runtime": 524.0268, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 1.3003454208374023, | |
| "learning_rate": 4.644547656614592e-05, | |
| "loss": 0.484, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 0.5635731816291809, | |
| "eval_runtime": 523.978, | |
| "eval_samples_per_second": 12.063, | |
| "eval_steps_per_second": 1.51, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 1.1711817979812622, | |
| "learning_rate": 4.622324940554235e-05, | |
| "loss": 0.4377, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 0.5571883916854858, | |
| "eval_runtime": 524.2962, | |
| "eval_samples_per_second": 12.056, | |
| "eval_steps_per_second": 1.509, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 1.191895842552185, | |
| "learning_rate": 4.6001022244938776e-05, | |
| "loss": 0.427, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.5517956614494324, | |
| "eval_runtime": 524.0245, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.1573776006698608, | |
| "learning_rate": 4.577879508433521e-05, | |
| "loss": 0.4216, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 0.5458714962005615, | |
| "eval_runtime": 523.9381, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 1.232151985168457, | |
| "learning_rate": 4.555656792373164e-05, | |
| "loss": 0.4272, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 0.5407230257987976, | |
| "eval_runtime": 523.9289, | |
| "eval_samples_per_second": 12.065, | |
| "eval_steps_per_second": 1.51, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 1.2596906423568726, | |
| "learning_rate": 4.5334340763128066e-05, | |
| "loss": 0.4197, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.5343000888824463, | |
| "eval_runtime": 524.1586, | |
| "eval_samples_per_second": 12.059, | |
| "eval_steps_per_second": 1.509, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 1.1676970720291138, | |
| "learning_rate": 4.51121136025245e-05, | |
| "loss": 0.4188, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 0.5275489687919617, | |
| "eval_runtime": 523.8327, | |
| "eval_samples_per_second": 12.067, | |
| "eval_steps_per_second": 1.51, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 1.291373372077942, | |
| "learning_rate": 4.4889886441920935e-05, | |
| "loss": 0.4165, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 0.5239323377609253, | |
| "eval_runtime": 523.8294, | |
| "eval_samples_per_second": 12.067, | |
| "eval_steps_per_second": 1.51, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 1.3542147874832153, | |
| "learning_rate": 4.466765928131736e-05, | |
| "loss": 0.411, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.5183799862861633, | |
| "eval_runtime": 523.956, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 1.138175368309021, | |
| "learning_rate": 4.4445432120713797e-05, | |
| "loss": 0.4066, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 0.5129222869873047, | |
| "eval_runtime": 524.4918, | |
| "eval_samples_per_second": 12.052, | |
| "eval_steps_per_second": 1.508, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 1.2202116250991821, | |
| "learning_rate": 4.4223204960110224e-05, | |
| "loss": 0.4054, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.5082316398620605, | |
| "eval_runtime": 524.0745, | |
| "eval_samples_per_second": 12.061, | |
| "eval_steps_per_second": 1.509, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 1.266581654548645, | |
| "learning_rate": 4.400097779950666e-05, | |
| "loss": 0.4015, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 0.5040789842605591, | |
| "eval_runtime": 524.5202, | |
| "eval_samples_per_second": 12.051, | |
| "eval_steps_per_second": 1.508, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 1.27601957321167, | |
| "learning_rate": 4.3778750638903086e-05, | |
| "loss": 0.3982, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 0.5011813044548035, | |
| "eval_runtime": 524.275, | |
| "eval_samples_per_second": 12.057, | |
| "eval_steps_per_second": 1.509, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 1.2219188213348389, | |
| "learning_rate": 4.355652347829952e-05, | |
| "loss": 0.3954, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_loss": 0.49553418159484863, | |
| "eval_runtime": 524.2405, | |
| "eval_samples_per_second": 12.057, | |
| "eval_steps_per_second": 1.509, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 1.0724387168884277, | |
| "learning_rate": 4.333429631769595e-05, | |
| "loss": 0.4003, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 0.4928473234176636, | |
| "eval_runtime": 524.0586, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.1879562139511108, | |
| "learning_rate": 4.311206915709238e-05, | |
| "loss": 0.3925, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.4879492521286011, | |
| "eval_runtime": 523.9744, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.207940697669983, | |
| "learning_rate": 4.288984199648881e-05, | |
| "loss": 0.3934, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 0.48524412512779236, | |
| "eval_runtime": 523.9417, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 1.1332272291183472, | |
| "learning_rate": 4.2667614835885244e-05, | |
| "loss": 0.3432, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 0.4873977303504944, | |
| "eval_runtime": 523.9543, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 1.1995199918746948, | |
| "learning_rate": 4.244538767528167e-05, | |
| "loss": 0.3242, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_loss": 0.48586371541023254, | |
| "eval_runtime": 524.0976, | |
| "eval_samples_per_second": 12.061, | |
| "eval_steps_per_second": 1.509, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 1.3645168542861938, | |
| "learning_rate": 4.2223160514678106e-05, | |
| "loss": 0.3249, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_loss": 0.4856572449207306, | |
| "eval_runtime": 525.0728, | |
| "eval_samples_per_second": 12.038, | |
| "eval_steps_per_second": 1.506, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 1.2310686111450195, | |
| "learning_rate": 4.200093335407454e-05, | |
| "loss": 0.3269, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_loss": 0.4839836359024048, | |
| "eval_runtime": 524.2734, | |
| "eval_samples_per_second": 12.057, | |
| "eval_steps_per_second": 1.509, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 1.1063857078552246, | |
| "learning_rate": 4.177870619347097e-05, | |
| "loss": 0.3208, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_loss": 0.4794396460056305, | |
| "eval_runtime": 524.0289, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 1.259097695350647, | |
| "learning_rate": 4.15564790328674e-05, | |
| "loss": 0.3285, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_loss": 0.4764421582221985, | |
| "eval_runtime": 523.9713, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 1.4079738855361938, | |
| "learning_rate": 4.133425187226383e-05, | |
| "loss": 0.3298, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_loss": 0.4742029011249542, | |
| "eval_runtime": 525.3854, | |
| "eval_samples_per_second": 12.031, | |
| "eval_steps_per_second": 1.506, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 1.3066357374191284, | |
| "learning_rate": 4.1112024711660265e-05, | |
| "loss": 0.3274, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_loss": 0.4709968864917755, | |
| "eval_runtime": 524.0046, | |
| "eval_samples_per_second": 12.063, | |
| "eval_steps_per_second": 1.51, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 1.236858606338501, | |
| "learning_rate": 4.088979755105669e-05, | |
| "loss": 0.3279, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_loss": 0.4686913788318634, | |
| "eval_runtime": 524.0329, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.256058931350708, | |
| "learning_rate": 4.066757039045313e-05, | |
| "loss": 0.3278, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_loss": 0.46550899744033813, | |
| "eval_runtime": 523.9744, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 1.51, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 1.3171123266220093, | |
| "learning_rate": 4.0445343229849554e-05, | |
| "loss": 0.325, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "eval_loss": 0.46436843276023865, | |
| "eval_runtime": 524.213, | |
| "eval_samples_per_second": 12.058, | |
| "eval_steps_per_second": 1.509, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 1.095387578010559, | |
| "learning_rate": 4.022311606924599e-05, | |
| "loss": 0.329, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_loss": 0.4598775804042816, | |
| "eval_runtime": 524.0933, | |
| "eval_samples_per_second": 12.061, | |
| "eval_steps_per_second": 1.509, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 1.1329820156097412, | |
| "learning_rate": 4.0000888908642416e-05, | |
| "loss": 0.3281, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_loss": 0.4569958746433258, | |
| "eval_runtime": 524.501, | |
| "eval_samples_per_second": 12.051, | |
| "eval_steps_per_second": 1.508, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 1.1984633207321167, | |
| "learning_rate": 3.977866174803885e-05, | |
| "loss": 0.3301, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_loss": 0.4568074643611908, | |
| "eval_runtime": 524.1689, | |
| "eval_samples_per_second": 12.059, | |
| "eval_steps_per_second": 1.509, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 1.2546519041061401, | |
| "learning_rate": 3.955643458743528e-05, | |
| "loss": 0.3329, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 0.4547139108181, | |
| "eval_runtime": 523.8736, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 1.51, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 1.077413558959961, | |
| "learning_rate": 3.933420742683171e-05, | |
| "loss": 0.3252, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_loss": 0.45176568627357483, | |
| "eval_runtime": 523.866, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 1.51, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 1.446174144744873, | |
| "learning_rate": 3.911198026622814e-05, | |
| "loss": 0.2923, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 0.46238696575164795, | |
| "eval_runtime": 524.0417, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 1.3876376152038574, | |
| "learning_rate": 3.8889753105624575e-05, | |
| "loss": 0.2567, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "eval_loss": 0.46406450867652893, | |
| "eval_runtime": 523.8474, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 1.51, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 1.3354049921035767, | |
| "learning_rate": 3.8667525945021e-05, | |
| "loss": 0.2639, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "eval_loss": 0.46170929074287415, | |
| "eval_runtime": 523.6743, | |
| "eval_samples_per_second": 12.07, | |
| "eval_steps_per_second": 1.51, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 1.3357325792312622, | |
| "learning_rate": 3.844529878441743e-05, | |
| "loss": 0.2679, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.462502658367157, | |
| "eval_runtime": 523.7861, | |
| "eval_samples_per_second": 12.068, | |
| "eval_steps_per_second": 1.51, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "grad_norm": 1.2143915891647339, | |
| "learning_rate": 3.8223071623813864e-05, | |
| "loss": 0.2678, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_loss": 0.461648166179657, | |
| "eval_runtime": 523.8835, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 1.51, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "grad_norm": 1.4215404987335205, | |
| "learning_rate": 3.800084446321029e-05, | |
| "loss": 0.272, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "eval_loss": 0.45769479870796204, | |
| "eval_runtime": 524.0212, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 1.258699893951416, | |
| "learning_rate": 3.7778617302606726e-05, | |
| "loss": 0.2734, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_loss": 0.45607540011405945, | |
| "eval_runtime": 523.6473, | |
| "eval_samples_per_second": 12.071, | |
| "eval_steps_per_second": 1.511, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "grad_norm": 1.329528570175171, | |
| "learning_rate": 3.7556390142003154e-05, | |
| "loss": 0.2744, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "eval_loss": 0.4543399214744568, | |
| "eval_runtime": 523.8153, | |
| "eval_samples_per_second": 12.067, | |
| "eval_steps_per_second": 1.51, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "grad_norm": 1.265796184539795, | |
| "learning_rate": 3.733416298139959e-05, | |
| "loss": 0.2727, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_loss": 0.4526574909687042, | |
| "eval_runtime": 523.7836, | |
| "eval_samples_per_second": 12.068, | |
| "eval_steps_per_second": 1.51, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "grad_norm": 1.3111358880996704, | |
| "learning_rate": 3.7111935820796016e-05, | |
| "loss": 0.2723, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_loss": 0.452004998922348, | |
| "eval_runtime": 523.8302, | |
| "eval_samples_per_second": 12.067, | |
| "eval_steps_per_second": 1.51, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 1.2918260097503662, | |
| "learning_rate": 3.688970866019245e-05, | |
| "loss": 0.2727, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "eval_loss": 0.4493946433067322, | |
| "eval_runtime": 523.8682, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 1.51, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 1.3306081295013428, | |
| "learning_rate": 3.666748149958888e-05, | |
| "loss": 0.2759, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "eval_loss": 0.4486713707447052, | |
| "eval_runtime": 524.0531, | |
| "eval_samples_per_second": 12.062, | |
| "eval_steps_per_second": 1.509, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "grad_norm": 1.253534197807312, | |
| "learning_rate": 3.644525433898531e-05, | |
| "loss": 0.2727, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 0.44542327523231506, | |
| "eval_runtime": 524.1367, | |
| "eval_samples_per_second": 12.06, | |
| "eval_steps_per_second": 1.509, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 1.1979799270629883, | |
| "learning_rate": 3.622302717838174e-05, | |
| "loss": 0.2747, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_loss": 0.44225621223449707, | |
| "eval_runtime": 523.5918, | |
| "eval_samples_per_second": 12.072, | |
| "eval_steps_per_second": 1.511, | |
| "step": 12400 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 45000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 14, | |
| "save_steps": 400, | |
| "total_flos": 8.769484235678417e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |