| { | |
| "best_global_step": 4500, | |
| "best_metric": 4.731945037841797, | |
| "best_model_checkpoint": "checkpoints/checkpoint-4500", | |
| "epoch": 4.999113362887904, | |
| "eval_steps": 500, | |
| "global_step": 4930, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1013299556681444, | |
| "grad_norm": 4.671788692474365, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 9.818, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2026599113362888, | |
| "grad_norm": 2.0833730697631836, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 6.0125, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3039898670044332, | |
| "grad_norm": 2.2255606651306152, | |
| "learning_rate": 2.96e-05, | |
| "loss": 5.5478, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4053198226725776, | |
| "grad_norm": 2.375220537185669, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 5.3821, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.506649778340722, | |
| "grad_norm": 2.1432723999023438, | |
| "learning_rate": 4.96e-05, | |
| "loss": 5.2798, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.506649778340722, | |
| "eval_loss": 5.099383354187012, | |
| "eval_runtime": 61.4753, | |
| "eval_samples_per_second": 57.08, | |
| "eval_steps_per_second": 14.282, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6079797340088664, | |
| "grad_norm": 2.5531654357910156, | |
| "learning_rate": 4.891647855530474e-05, | |
| "loss": 5.2792, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7093096896770108, | |
| "grad_norm": 1.8546665906906128, | |
| "learning_rate": 4.7787810383747176e-05, | |
| "loss": 5.1646, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8106396453451552, | |
| "grad_norm": 2.103972911834717, | |
| "learning_rate": 4.665914221218962e-05, | |
| "loss": 5.1678, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.9119696010132996, | |
| "grad_norm": 2.0656638145446777, | |
| "learning_rate": 4.553047404063205e-05, | |
| "loss": 5.1145, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.0141861937935401, | |
| "grad_norm": 2.003814697265625, | |
| "learning_rate": 4.440180586907449e-05, | |
| "loss": 5.1537, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0141861937935401, | |
| "eval_loss": 4.938778877258301, | |
| "eval_runtime": 61.4905, | |
| "eval_samples_per_second": 57.066, | |
| "eval_steps_per_second": 14.279, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.1155161494616845, | |
| "grad_norm": 2.0927860736846924, | |
| "learning_rate": 4.327313769751693e-05, | |
| "loss": 5.0454, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.216846105129829, | |
| "grad_norm": 2.080012321472168, | |
| "learning_rate": 4.214446952595937e-05, | |
| "loss": 5.0665, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.3181760607979733, | |
| "grad_norm": 1.8747535943984985, | |
| "learning_rate": 4.101580135440181e-05, | |
| "loss": 4.9706, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.4195060164661177, | |
| "grad_norm": 1.901370882987976, | |
| "learning_rate": 3.988713318284424e-05, | |
| "loss": 4.9639, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.5208359721342621, | |
| "grad_norm": 1.9110698699951172, | |
| "learning_rate": 3.875846501128668e-05, | |
| "loss": 4.9622, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5208359721342621, | |
| "eval_loss": 4.870736598968506, | |
| "eval_runtime": 61.3393, | |
| "eval_samples_per_second": 57.206, | |
| "eval_steps_per_second": 14.314, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.6221659278024065, | |
| "grad_norm": 1.8562686443328857, | |
| "learning_rate": 3.762979683972912e-05, | |
| "loss": 4.9633, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.723495883470551, | |
| "grad_norm": 1.970841884613037, | |
| "learning_rate": 3.650112866817156e-05, | |
| "loss": 4.9279, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.8248258391386953, | |
| "grad_norm": 1.9571095705032349, | |
| "learning_rate": 3.5372460496614e-05, | |
| "loss": 4.952, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.9261557948068397, | |
| "grad_norm": 2.0035080909729004, | |
| "learning_rate": 3.424379232505643e-05, | |
| "loss": 4.9448, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.0283723875870803, | |
| "grad_norm": 1.978408694267273, | |
| "learning_rate": 3.3115124153498873e-05, | |
| "loss": 4.9862, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0283723875870803, | |
| "eval_loss": 4.826657295227051, | |
| "eval_runtime": 61.364, | |
| "eval_samples_per_second": 57.183, | |
| "eval_steps_per_second": 14.308, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.1297023432552247, | |
| "grad_norm": 2.2265381813049316, | |
| "learning_rate": 3.198645598194131e-05, | |
| "loss": 4.8837, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.231032298923369, | |
| "grad_norm": 1.8263903856277466, | |
| "learning_rate": 3.085778781038375e-05, | |
| "loss": 4.893, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.3323622545915135, | |
| "grad_norm": 1.8423362970352173, | |
| "learning_rate": 2.9729119638826186e-05, | |
| "loss": 4.9038, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.433692210259658, | |
| "grad_norm": 1.9064007997512817, | |
| "learning_rate": 2.8600451467268623e-05, | |
| "loss": 4.8971, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.5350221659278023, | |
| "grad_norm": 1.8199445009231567, | |
| "learning_rate": 2.747178329571106e-05, | |
| "loss": 4.8563, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.5350221659278023, | |
| "eval_loss": 4.789968013763428, | |
| "eval_runtime": 61.2645, | |
| "eval_samples_per_second": 57.276, | |
| "eval_steps_per_second": 14.331, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.6363521215959467, | |
| "grad_norm": 2.2458302974700928, | |
| "learning_rate": 2.63431151241535e-05, | |
| "loss": 4.8214, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.737682077264091, | |
| "grad_norm": 1.9292908906936646, | |
| "learning_rate": 2.521444695259594e-05, | |
| "loss": 4.8538, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.8390120329322355, | |
| "grad_norm": 2.03075909614563, | |
| "learning_rate": 2.4085778781038376e-05, | |
| "loss": 4.8307, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.94034198860038, | |
| "grad_norm": 1.909643530845642, | |
| "learning_rate": 2.2957110609480814e-05, | |
| "loss": 4.8282, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.0425585813806206, | |
| "grad_norm": 1.8887925148010254, | |
| "learning_rate": 2.182844243792325e-05, | |
| "loss": 4.8738, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0425585813806206, | |
| "eval_loss": 4.765661716461182, | |
| "eval_runtime": 61.3376, | |
| "eval_samples_per_second": 57.208, | |
| "eval_steps_per_second": 14.314, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.143888537048765, | |
| "grad_norm": 1.8953306674957275, | |
| "learning_rate": 2.069977426636569e-05, | |
| "loss": 4.7451, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.2452184927169094, | |
| "grad_norm": 2.1468937397003174, | |
| "learning_rate": 1.957110609480813e-05, | |
| "loss": 4.7813, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.346548448385054, | |
| "grad_norm": 1.9347341060638428, | |
| "learning_rate": 1.8442437923250567e-05, | |
| "loss": 4.792, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.4478784040531982, | |
| "grad_norm": 1.8998669385910034, | |
| "learning_rate": 1.7313769751693004e-05, | |
| "loss": 4.7867, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.5492083597213426, | |
| "grad_norm": 1.899141788482666, | |
| "learning_rate": 1.6185101580135442e-05, | |
| "loss": 4.8095, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.5492083597213426, | |
| "eval_loss": 4.752286434173584, | |
| "eval_runtime": 61.3687, | |
| "eval_samples_per_second": 57.179, | |
| "eval_steps_per_second": 14.307, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.650538315389487, | |
| "grad_norm": 1.9524105787277222, | |
| "learning_rate": 1.5056433408577881e-05, | |
| "loss": 4.8354, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.7518682710576314, | |
| "grad_norm": 2.0022027492523193, | |
| "learning_rate": 1.3927765237020315e-05, | |
| "loss": 4.8232, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.853198226725776, | |
| "grad_norm": 1.9039005041122437, | |
| "learning_rate": 1.2799097065462754e-05, | |
| "loss": 4.7974, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.9545281823939202, | |
| "grad_norm": 1.902718186378479, | |
| "learning_rate": 1.1670428893905193e-05, | |
| "loss": 4.8301, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.0567447751741605, | |
| "grad_norm": 2.1200180053710938, | |
| "learning_rate": 1.054176072234763e-05, | |
| "loss": 4.8426, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0567447751741605, | |
| "eval_loss": 4.738107681274414, | |
| "eval_runtime": 61.3616, | |
| "eval_samples_per_second": 57.186, | |
| "eval_steps_per_second": 14.309, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.158074730842305, | |
| "grad_norm": 2.008924961090088, | |
| "learning_rate": 9.413092550790068e-06, | |
| "loss": 4.7473, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 4.259404686510449, | |
| "grad_norm": 1.881294846534729, | |
| "learning_rate": 8.284424379232506e-06, | |
| "loss": 4.7348, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.360734642178594, | |
| "grad_norm": 1.8294726610183716, | |
| "learning_rate": 7.155756207674943e-06, | |
| "loss": 4.7728, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 4.462064597846738, | |
| "grad_norm": 1.9369500875473022, | |
| "learning_rate": 6.0270880361173815e-06, | |
| "loss": 4.7811, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.5633945535148825, | |
| "grad_norm": 1.9159400463104248, | |
| "learning_rate": 4.89841986455982e-06, | |
| "loss": 4.7478, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.5633945535148825, | |
| "eval_loss": 4.731945037841797, | |
| "eval_runtime": 61.3768, | |
| "eval_samples_per_second": 57.171, | |
| "eval_steps_per_second": 14.305, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.664724509183027, | |
| "grad_norm": 1.9592151641845703, | |
| "learning_rate": 3.7697516930022577e-06, | |
| "loss": 4.7561, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.766054464851171, | |
| "grad_norm": 1.8825119733810425, | |
| "learning_rate": 2.6410835214446955e-06, | |
| "loss": 4.8118, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 4.867384420519316, | |
| "grad_norm": 2.025451898574829, | |
| "learning_rate": 1.5124153498871334e-06, | |
| "loss": 4.7582, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.96871437618746, | |
| "grad_norm": 2.0353872776031494, | |
| "learning_rate": 3.837471783295711e-07, | |
| "loss": 4.7274, | |
| "step": 4900 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4930, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 8.1592623788544e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |