| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.999089792017476, | |
| "eval_steps": 500, | |
| "global_step": 5492, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18204159650480134, | |
| "grad_norm": 3.854499101638794, | |
| "learning_rate": 9.548269581056467e-05, | |
| "loss": 1.0014, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3640831930096027, | |
| "grad_norm": 3.5418713092803955, | |
| "learning_rate": 9.092896174863389e-05, | |
| "loss": 0.5401, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.546124789514404, | |
| "grad_norm": 3.089492082595825, | |
| "learning_rate": 8.63752276867031e-05, | |
| "loss": 0.4797, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.7281663860192054, | |
| "grad_norm": 3.0062525272369385, | |
| "learning_rate": 8.182149362477231e-05, | |
| "loss": 0.4598, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9102079825240067, | |
| "grad_norm": 2.703545093536377, | |
| "learning_rate": 7.726775956284153e-05, | |
| "loss": 0.4454, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.092249579028808, | |
| "grad_norm": 2.665408134460449, | |
| "learning_rate": 7.271402550091076e-05, | |
| "loss": 0.4338, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.2742911755336095, | |
| "grad_norm": 2.8605291843414307, | |
| "learning_rate": 6.816029143897996e-05, | |
| "loss": 0.4224, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.4563327720384107, | |
| "grad_norm": 3.0206186771392822, | |
| "learning_rate": 6.360655737704918e-05, | |
| "loss": 0.4147, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6383743685432122, | |
| "grad_norm": 3.633958578109741, | |
| "learning_rate": 5.90528233151184e-05, | |
| "loss": 0.3983, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.8204159650480136, | |
| "grad_norm": 3.535999059677124, | |
| "learning_rate": 5.449908925318762e-05, | |
| "loss": 0.3808, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0024575615528146, | |
| "grad_norm": 3.919354200363159, | |
| "learning_rate": 4.994535519125683e-05, | |
| "loss": 0.3642, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.184499158057616, | |
| "grad_norm": 4.2734246253967285, | |
| "learning_rate": 4.539162112932605e-05, | |
| "loss": 0.3377, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.3665407545624175, | |
| "grad_norm": 4.275780200958252, | |
| "learning_rate": 4.083788706739526e-05, | |
| "loss": 0.3261, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.548582351067219, | |
| "grad_norm": 4.408421516418457, | |
| "learning_rate": 3.6284153005464486e-05, | |
| "loss": 0.31, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.7306239475720204, | |
| "grad_norm": 4.726068019866943, | |
| "learning_rate": 3.17304189435337e-05, | |
| "loss": 0.2998, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.9126655440768214, | |
| "grad_norm": 4.753103733062744, | |
| "learning_rate": 2.7176684881602916e-05, | |
| "loss": 0.2897, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.094707140581623, | |
| "grad_norm": 4.001144886016846, | |
| "learning_rate": 2.262295081967213e-05, | |
| "loss": 0.2744, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 3.2767487370864243, | |
| "grad_norm": 4.412220001220703, | |
| "learning_rate": 1.806921675774135e-05, | |
| "loss": 0.2619, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.458790333591226, | |
| "grad_norm": 4.553833961486816, | |
| "learning_rate": 1.3515482695810567e-05, | |
| "loss": 0.2578, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 3.640831930096027, | |
| "grad_norm": 4.323617458343506, | |
| "learning_rate": 8.961748633879782e-06, | |
| "loss": 0.2533, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.8228735266008282, | |
| "grad_norm": 4.197403430938721, | |
| "learning_rate": 4.408014571948998e-06, | |
| "loss": 0.2489, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 3.999089792017476, | |
| "step": 5492, | |
| "total_flos": 1.2115246643852867e+18, | |
| "train_loss": 0.38415815321442337, | |
| "train_runtime": 38374.2384, | |
| "train_samples_per_second": 13.742, | |
| "train_steps_per_second": 0.143 | |
| } | |
| ], | |
| "logging_steps": 250, | |
| "max_steps": 5492, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 1.2115246643852867e+18, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |