| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 2907, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03439972480220158, | |
| "grad_norm": 5.509068012237549, | |
| "learning_rate": 8.60215053763441e-07, | |
| "loss": 6.3428, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06879944960440317, | |
| "grad_norm": 5.2295918464660645, | |
| "learning_rate": 1.720430107526882e-06, | |
| "loss": 6.2525, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.10319917440660474, | |
| "grad_norm": 4.567111492156982, | |
| "learning_rate": 2.580645161290323e-06, | |
| "loss": 5.9325, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13759889920880633, | |
| "grad_norm": 3.807996988296509, | |
| "learning_rate": 3.440860215053764e-06, | |
| "loss": 5.2401, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1719986240110079, | |
| "grad_norm": 3.0099501609802246, | |
| "learning_rate": 4.3010752688172045e-06, | |
| "loss": 4.5705, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.20639834881320948, | |
| "grad_norm": 1.4484069347381592, | |
| "learning_rate": 5.161290322580646e-06, | |
| "loss": 4.253, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.24079807361541108, | |
| "grad_norm": 1.3093016147613525, | |
| "learning_rate": 6.021505376344087e-06, | |
| "loss": 4.0894, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.27519779841761266, | |
| "grad_norm": 1.5757373571395874, | |
| "learning_rate": 6.881720430107528e-06, | |
| "loss": 3.9769, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.30959752321981426, | |
| "grad_norm": 1.3851218223571777, | |
| "learning_rate": 7.741935483870968e-06, | |
| "loss": 3.9085, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3439972480220158, | |
| "grad_norm": 1.577419638633728, | |
| "learning_rate": 8.602150537634409e-06, | |
| "loss": 3.8505, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3783969728242174, | |
| "grad_norm": 1.839663028717041, | |
| "learning_rate": 9.46236559139785e-06, | |
| "loss": 3.8197, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.41279669762641896, | |
| "grad_norm": 26.901336669921875, | |
| "learning_rate": 1.0322580645161291e-05, | |
| "loss": 3.7087, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.44719642242862057, | |
| "grad_norm": 2.2534244060516357, | |
| "learning_rate": 1.118279569892473e-05, | |
| "loss": 3.6651, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.48159614723082217, | |
| "grad_norm": 3.677131175994873, | |
| "learning_rate": 1.2043010752688173e-05, | |
| "loss": 3.5098, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5159958720330238, | |
| "grad_norm": 3.7944538593292236, | |
| "learning_rate": 1.2903225806451613e-05, | |
| "loss": 3.3414, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5503955968352253, | |
| "grad_norm": 4.084872722625732, | |
| "learning_rate": 1.3763440860215056e-05, | |
| "loss": 3.1296, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.5847953216374269, | |
| "grad_norm": 2.373960494995117, | |
| "learning_rate": 1.4623655913978497e-05, | |
| "loss": 2.8209, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.6191950464396285, | |
| "grad_norm": 2.3671751022338867, | |
| "learning_rate": 1.5483870967741936e-05, | |
| "loss": 2.5835, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.6535947712418301, | |
| "grad_norm": 3.086862564086914, | |
| "learning_rate": 1.6344086021505377e-05, | |
| "loss": 2.4355, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.6879944960440316, | |
| "grad_norm": 2.4583280086517334, | |
| "learning_rate": 1.7204301075268818e-05, | |
| "loss": 2.4215, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7223942208462333, | |
| "grad_norm": 1.4451029300689697, | |
| "learning_rate": 1.806451612903226e-05, | |
| "loss": 2.2917, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.7567939456484348, | |
| "grad_norm": 1.6850578784942627, | |
| "learning_rate": 1.89247311827957e-05, | |
| "loss": 2.2604, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.7911936704506364, | |
| "grad_norm": 2.9293603897094727, | |
| "learning_rate": 1.978494623655914e-05, | |
| "loss": 2.2541, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.8255933952528379, | |
| "grad_norm": 1.4160730838775635, | |
| "learning_rate": 1.9191636897958123e-05, | |
| "loss": 2.2484, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.8599931200550396, | |
| "grad_norm": 1.52695631980896, | |
| "learning_rate": 1.586037078125607e-05, | |
| "loss": 2.177, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8943928448572411, | |
| "grad_norm": 1.5154156684875488, | |
| "learning_rate": 1.086259476800041e-05, | |
| "loss": 2.1612, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.9287925696594427, | |
| "grad_norm": 2.4500575065612793, | |
| "learning_rate": 5.619523231433177e-06, | |
| "loss": 2.156, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.9631922944616443, | |
| "grad_norm": 1.812966227531433, | |
| "learning_rate": 1.6221250761114803e-06, | |
| "loss": 2.163, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.9975920192638459, | |
| "grad_norm": 1.3679808378219604, | |
| "learning_rate": 7.1378626715268295e-09, | |
| "loss": 2.1348, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 2907, | |
| "total_flos": 3.458455817609544e+17, | |
| "train_loss": 3.4343306545051546, | |
| "train_runtime": 789.0664, | |
| "train_samples_per_second": 29.473, | |
| "train_steps_per_second": 3.684 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2907, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.458455817609544e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |