| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "global_step": 44145, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00029759999999999997, | |
| "loss": 2.9274, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00029660442204147095, | |
| "loss": 0.9437, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002931676022453889, | |
| "loss": 0.8203, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0002897307824493069, | |
| "loss": 0.7688, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0002862939626532249, | |
| "loss": 0.7259, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002828571428571428, | |
| "loss": 0.6873, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0002794203230610608, | |
| "loss": 0.6257, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.0002759835032649788, | |
| "loss": 0.6302, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00027255355710848893, | |
| "loss": 0.6274, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.0002691167373124069, | |
| "loss": 0.6076, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00026567991751632485, | |
| "loss": 0.5965, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00026224309772024287, | |
| "loss": 0.5949, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.00025880627792416083, | |
| "loss": 0.543, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0002553694581280788, | |
| "loss": 0.53, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.00025193263833199676, | |
| "loss": 0.5233, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 0.0002484958185359147, | |
| "loss": 0.5354, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 0.00024505899873983274, | |
| "loss": 0.5302, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.00024162905258334286, | |
| "loss": 0.5072, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00023819223278726085, | |
| "loss": 0.4687, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 0.00023475541299117878, | |
| "loss": 0.4736, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_loss": 0.624671459197998, | |
| "eval_runtime": 84.9387, | |
| "eval_samples_per_second": 68.108, | |
| "eval_steps_per_second": 4.262, | |
| "eval_wer": 0.6361829025844931, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 0.00023131859319509677, | |
| "loss": 0.4707, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 0.00022788864703860692, | |
| "loss": 0.4712, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.00022445182724252488, | |
| "loss": 0.4596, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 0.00022101500744644287, | |
| "loss": 0.4394, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.00021757818765036084, | |
| "loss": 0.4136, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 0.00021414136785427883, | |
| "loss": 0.4233, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.00021071142169778895, | |
| "loss": 0.4353, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.00020727460190170694, | |
| "loss": 0.4248, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 0.00020383778210562493, | |
| "loss": 0.418, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 0.0002004009623095429, | |
| "loss": 0.3875, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.00019696414251346088, | |
| "loss": 0.3812, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.0001935273227173788, | |
| "loss": 0.3817, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 0.0001900905029212968, | |
| "loss": 0.3821, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.0001866536831252148, | |
| "loss": 0.3684, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "learning_rate": 0.00018321686332913275, | |
| "loss": 0.3791, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 0.00017978004353305074, | |
| "loss": 0.3536, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.00017634322373696873, | |
| "loss": 0.33, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 0.00017290640394088667, | |
| "loss": 0.3324, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 0.00016946958414480466, | |
| "loss": 0.3315, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.00016603963798831478, | |
| "loss": 0.3392, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_loss": 0.7254078388214111, | |
| "eval_runtime": 84.9657, | |
| "eval_samples_per_second": 68.086, | |
| "eval_steps_per_second": 4.261, | |
| "eval_wer": 0.5604705102717031, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 0.00016260281819223277, | |
| "loss": 0.3343, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 0.00015917974567533508, | |
| "loss": 0.3206, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 0.00015574292587925307, | |
| "loss": 0.3032, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 0.00015230610608317103, | |
| "loss": 0.3005, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 0.00014886928628708902, | |
| "loss": 0.3025, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.82, | |
| "learning_rate": 0.00014543246649100698, | |
| "loss": 0.3011, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 0.00014200252033451713, | |
| "loss": 0.2985, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 0.0001385657005384351, | |
| "loss": 0.2646, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 0.00013512888074235306, | |
| "loss": 0.2726, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 0.00013169206094627105, | |
| "loss": 0.2647, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.000128255241150189, | |
| "loss": 0.2722, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 0.000124818421354107, | |
| "loss": 0.2637, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 0.00012138160155802496, | |
| "loss": 0.2593, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 0.00011794478176194294, | |
| "loss": 0.2332, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "learning_rate": 0.00011451483560545307, | |
| "loss": 0.2425, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "learning_rate": 0.00011107801580937105, | |
| "loss": 0.2392, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.00010764119601328904, | |
| "loss": 0.2327, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 0.000104204376217207, | |
| "loss": 0.2305, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 0.00010077443006071715, | |
| "loss": 0.2268, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 9.733761026463512e-05, | |
| "loss": 0.2085, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "eval_loss": 0.5465497970581055, | |
| "eval_runtime": 84.7941, | |
| "eval_samples_per_second": 68.224, | |
| "eval_steps_per_second": 4.269, | |
| "eval_wer": 0.5096918489065606, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 10.36, | |
| "learning_rate": 9.390079046855309e-05, | |
| "loss": 0.2124, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "learning_rate": 9.046397067247106e-05, | |
| "loss": 0.2078, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 8.702715087638905e-05, | |
| "loss": 0.2046, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 8.359033108030701e-05, | |
| "loss": 0.2055, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 8.015351128422499e-05, | |
| "loss": 0.2009, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "learning_rate": 7.671669148814298e-05, | |
| "loss": 0.179, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 7.327987169206094e-05, | |
| "loss": 0.1815, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 11.55, | |
| "learning_rate": 6.984305189597892e-05, | |
| "loss": 0.1835, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 6.640623209989689e-05, | |
| "loss": 0.1819, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 6.296941230381487e-05, | |
| "loss": 0.1771, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 5.953259250773284e-05, | |
| "loss": 0.1708, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 5.609577271165081e-05, | |
| "loss": 0.16, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 5.265895291556879e-05, | |
| "loss": 0.1643, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 12.57, | |
| "learning_rate": 4.9222133119486765e-05, | |
| "loss": 0.1578, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 4.578531332340474e-05, | |
| "loss": 0.1521, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 4.2348493527322716e-05, | |
| "loss": 0.1592, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "learning_rate": 3.8911673731240685e-05, | |
| "loss": 0.1453, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 13.25, | |
| "learning_rate": 3.548172757475083e-05, | |
| "loss": 0.1437, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "learning_rate": 3.2044907778668804e-05, | |
| "loss": 0.1353, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "learning_rate": 2.8614961622178944e-05, | |
| "loss": 0.1387, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "eval_loss": 0.598371684551239, | |
| "eval_runtime": 84.9323, | |
| "eval_samples_per_second": 68.113, | |
| "eval_steps_per_second": 4.262, | |
| "eval_wer": 0.46322067594433397, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 2.5178141826096916e-05, | |
| "loss": 0.1388, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 2.1741322030014892e-05, | |
| "loss": 0.1376, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 1.8304502233932868e-05, | |
| "loss": 0.137, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 14.27, | |
| "learning_rate": 1.486768243785084e-05, | |
| "loss": 0.1257, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "learning_rate": 1.1430862641768814e-05, | |
| "loss": 0.1243, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 14.61, | |
| "learning_rate": 7.99404284568679e-06, | |
| "loss": 0.1207, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 4.557223049604766e-06, | |
| "loss": 0.121, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 1.1272768931149044e-06, | |
| "loss": 0.1214, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 44145, | |
| "total_flos": 1.429924244829818e+19, | |
| "train_loss": 0.3679376628312752, | |
| "train_runtime": 9694.3418, | |
| "train_samples_per_second": 36.425, | |
| "train_steps_per_second": 4.554 | |
| } | |
| ], | |
| "max_steps": 44145, | |
| "num_train_epochs": 15, | |
| "total_flos": 1.429924244829818e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |