| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 3000, | |
| "global_step": 44343, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.76544212164265e-05, | |
| "grad_norm": 7.09375, | |
| "learning_rate": 0.0002999932345578783, | |
| "loss": 4.25, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.20296326364927947, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 0.000279703673635072, | |
| "loss": 2.7986, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.20296326364927947, | |
| "eval_loss": 2.3799755573272705, | |
| "eval_runtime": 61.2814, | |
| "eval_samples_per_second": 1540.516, | |
| "eval_steps_per_second": 6.021, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.40592652729855894, | |
| "grad_norm": 0.77734375, | |
| "learning_rate": 0.0002594073472701441, | |
| "loss": 2.7128, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.40592652729855894, | |
| "eval_loss": 2.3596713542938232, | |
| "eval_runtime": 61.2743, | |
| "eval_samples_per_second": 1540.694, | |
| "eval_steps_per_second": 6.022, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6088897909478385, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 0.0002391110209052161, | |
| "loss": 2.691, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.6088897909478385, | |
| "eval_loss": 2.348450183868408, | |
| "eval_runtime": 61.2066, | |
| "eval_samples_per_second": 1542.4, | |
| "eval_steps_per_second": 6.029, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8118530545971179, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 0.0002188146945402882, | |
| "loss": 2.6832, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.8118530545971179, | |
| "eval_loss": 2.3406801223754883, | |
| "eval_runtime": 61.3132, | |
| "eval_samples_per_second": 1539.719, | |
| "eval_steps_per_second": 6.018, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0148163182463974, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 0.00019851836817536025, | |
| "loss": 2.6763, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.0148163182463974, | |
| "eval_loss": 2.3388381004333496, | |
| "eval_runtime": 61.2522, | |
| "eval_samples_per_second": 1541.25, | |
| "eval_steps_per_second": 6.024, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.217779581895677, | |
| "grad_norm": 0.8359375, | |
| "learning_rate": 0.0001782220418104323, | |
| "loss": 2.6714, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.217779581895677, | |
| "eval_loss": 2.3343918323516846, | |
| "eval_runtime": 61.2808, | |
| "eval_samples_per_second": 1540.532, | |
| "eval_steps_per_second": 6.021, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.4207428455449564, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 0.00015792571544550436, | |
| "loss": 2.6693, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.4207428455449564, | |
| "eval_loss": 2.333566188812256, | |
| "eval_runtime": 61.2712, | |
| "eval_samples_per_second": 1540.773, | |
| "eval_steps_per_second": 6.022, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.6237061091942357, | |
| "grad_norm": 0.8359375, | |
| "learning_rate": 0.0001376293890805764, | |
| "loss": 2.6672, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.6237061091942357, | |
| "eval_loss": 2.3340320587158203, | |
| "eval_runtime": 61.1843, | |
| "eval_samples_per_second": 1542.961, | |
| "eval_steps_per_second": 6.031, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.8266693728435153, | |
| "grad_norm": 0.93359375, | |
| "learning_rate": 0.00011733306271564845, | |
| "loss": 2.6654, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.8266693728435153, | |
| "eval_loss": 2.3319995403289795, | |
| "eval_runtime": 61.2774, | |
| "eval_samples_per_second": 1540.618, | |
| "eval_steps_per_second": 6.022, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.029632636492795, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 9.703673635072052e-05, | |
| "loss": 2.6655, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.029632636492795, | |
| "eval_loss": 2.3317666053771973, | |
| "eval_runtime": 61.4137, | |
| "eval_samples_per_second": 1537.199, | |
| "eval_steps_per_second": 6.008, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.232595900142074, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 7.674040998579256e-05, | |
| "loss": 2.6628, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.232595900142074, | |
| "eval_loss": 2.3321053981781006, | |
| "eval_runtime": 61.2773, | |
| "eval_samples_per_second": 1540.618, | |
| "eval_steps_per_second": 6.022, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.435559163791354, | |
| "grad_norm": 0.765625, | |
| "learning_rate": 5.644408362086462e-05, | |
| "loss": 2.666, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.435559163791354, | |
| "eval_loss": 2.3318090438842773, | |
| "eval_runtime": 61.27, | |
| "eval_samples_per_second": 1540.804, | |
| "eval_steps_per_second": 6.023, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.638522427440633, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 3.614775725593667e-05, | |
| "loss": 2.6666, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.638522427440633, | |
| "eval_loss": 2.331491470336914, | |
| "eval_runtime": 61.2796, | |
| "eval_samples_per_second": 1540.562, | |
| "eval_steps_per_second": 6.022, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.841485691089913, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 1.5851430891008727e-05, | |
| "loss": 2.665, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.841485691089913, | |
| "eval_loss": 2.331724166870117, | |
| "eval_runtime": 61.1466, | |
| "eval_samples_per_second": 1543.912, | |
| "eval_steps_per_second": 6.035, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 44343, | |
| "total_flos": 4.6159075948098355e+17, | |
| "train_loss": 2.682002373542611, | |
| "train_runtime": 22758.1142, | |
| "train_samples_per_second": 498.788, | |
| "train_steps_per_second": 1.948 | |
| } | |
| ], | |
| "logging_steps": 3000, | |
| "max_steps": 44343, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 3000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.6159075948098355e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |