| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9957446808510637, | |
| "eval_steps": 500, | |
| "global_step": 264, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11347517730496454, | |
| "grad_norm": 2.358112635127941, | |
| "learning_rate": 5e-06, | |
| "loss": 1.045, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22695035460992907, | |
| "grad_norm": 2.2025871190711044, | |
| "learning_rate": 5e-06, | |
| "loss": 0.939, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3404255319148936, | |
| "grad_norm": 1.183824809408691, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8991, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.45390070921985815, | |
| "grad_norm": 1.2551021476147783, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8706, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5673758865248227, | |
| "grad_norm": 1.6978646589888085, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8565, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6808510638297872, | |
| "grad_norm": 2.13778699873673, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8386, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7943262411347518, | |
| "grad_norm": 0.993295806091264, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8309, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9078014184397163, | |
| "grad_norm": 1.1398762974593635, | |
| "learning_rate": 5e-06, | |
| "loss": 0.823, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9985815602836879, | |
| "eval_loss": 0.8090236783027649, | |
| "eval_runtime": 63.3465, | |
| "eval_samples_per_second": 37.445, | |
| "eval_steps_per_second": 0.6, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.0212765957446808, | |
| "grad_norm": 1.013500095467207, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8742, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1347517730496455, | |
| "grad_norm": 1.056225323349834, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7667, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.24822695035461, | |
| "grad_norm": 0.7290196034792423, | |
| "learning_rate": 5e-06, | |
| "loss": 0.755, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.3617021276595744, | |
| "grad_norm": 0.8838498260846974, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7554, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.475177304964539, | |
| "grad_norm": 0.821991213787815, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7556, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5886524822695036, | |
| "grad_norm": 0.9855152726966359, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7493, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.702127659574468, | |
| "grad_norm": 0.6490086567527167, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7477, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8156028368794326, | |
| "grad_norm": 1.0694149262660388, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7414, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.9290780141843973, | |
| "grad_norm": 0.9645011140855406, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7481, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.9971631205673759, | |
| "eval_loss": 0.7898643016815186, | |
| "eval_runtime": 62.2492, | |
| "eval_samples_per_second": 38.105, | |
| "eval_steps_per_second": 0.61, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 2.0425531914893615, | |
| "grad_norm": 1.581859190270789, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7818, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.1560283687943262, | |
| "grad_norm": 1.0466470957786433, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6863, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.269503546099291, | |
| "grad_norm": 0.9663026123669691, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6798, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.382978723404255, | |
| "grad_norm": 0.8243226264574698, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6826, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.49645390070922, | |
| "grad_norm": 1.0907354136557872, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6839, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.6099290780141846, | |
| "grad_norm": 0.7996806357479502, | |
| "learning_rate": 5e-06, | |
| "loss": 0.687, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.723404255319149, | |
| "grad_norm": 0.9108831837931511, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6902, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.8368794326241136, | |
| "grad_norm": 0.8473372600949097, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6873, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.950354609929078, | |
| "grad_norm": 0.8244777156304377, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6866, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.9957446808510637, | |
| "eval_loss": 0.7845782041549683, | |
| "eval_runtime": 58.8988, | |
| "eval_samples_per_second": 40.272, | |
| "eval_steps_per_second": 0.645, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 2.9957446808510637, | |
| "step": 264, | |
| "total_flos": 442000453140480.0, | |
| "train_loss": 0.7778021304896383, | |
| "train_runtime": 8882.1716, | |
| "train_samples_per_second": 15.22, | |
| "train_steps_per_second": 0.03 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 264, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 442000453140480.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |