{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.014871731317387532, "eval_steps": 5, "global_step": 15, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0009914487544925021, "grad_norm": 0.604952871799469, "learning_rate": 1e-05, "loss": 3.1854, "step": 1 }, { "epoch": 0.0009914487544925021, "eval_loss": 3.30363392829895, "eval_runtime": 78.7969, "eval_samples_per_second": 21.562, "eval_steps_per_second": 10.787, "step": 1 }, { "epoch": 0.0019828975089850043, "grad_norm": 0.5781750679016113, "learning_rate": 2e-05, "loss": 3.1779, "step": 2 }, { "epoch": 0.0029743462634775064, "grad_norm": 0.7077589631080627, "learning_rate": 3e-05, "loss": 3.4209, "step": 3 }, { "epoch": 0.0039657950179700086, "grad_norm": 0.6238883137702942, "learning_rate": 4e-05, "loss": 3.2321, "step": 4 }, { "epoch": 0.004957243772462511, "grad_norm": 0.6038368344306946, "learning_rate": 5e-05, "loss": 3.2133, "step": 5 }, { "epoch": 0.004957243772462511, "eval_loss": 3.2816126346588135, "eval_runtime": 7.5553, "eval_samples_per_second": 224.876, "eval_steps_per_second": 112.504, "step": 5 }, { "epoch": 0.005948692526955013, "grad_norm": 0.5721911787986755, "learning_rate": 6e-05, "loss": 3.1292, "step": 6 }, { "epoch": 0.006940141281447515, "grad_norm": 0.6283555030822754, "learning_rate": 7e-05, "loss": 3.1542, "step": 7 }, { "epoch": 0.007931590035940017, "grad_norm": 0.5923196077346802, "learning_rate": 8e-05, "loss": 3.2067, "step": 8 }, { "epoch": 0.00892303879043252, "grad_norm": 0.5854600667953491, "learning_rate": 9e-05, "loss": 3.1732, "step": 9 }, { "epoch": 0.009914487544925021, "grad_norm": 0.5889576077461243, "learning_rate": 0.0001, "loss": 3.1087, "step": 10 }, { "epoch": 0.009914487544925021, "eval_loss": 3.2038986682891846, "eval_runtime": 7.3532, "eval_samples_per_second": 231.054, "eval_steps_per_second": 115.595, "step": 10 }, { "epoch": 0.010905936299417524, "grad_norm": 0.6500622630119324, "learning_rate": 9.755282581475769e-05, "loss": 3.1351, "step": 11 }, { "epoch": 0.011897385053910026, "grad_norm": 0.5976778864860535, "learning_rate": 9.045084971874738e-05, "loss": 3.1733, "step": 12 }, { "epoch": 0.012888833808402528, "grad_norm": 0.6121130585670471, "learning_rate": 7.938926261462366e-05, "loss": 3.215, "step": 13 }, { "epoch": 0.01388028256289503, "grad_norm": 0.640393078327179, "learning_rate": 6.545084971874738e-05, "loss": 3.0551, "step": 14 }, { "epoch": 0.014871731317387532, "grad_norm": 0.5806251764297485, "learning_rate": 5e-05, "loss": 3.0478, "step": 15 }, { "epoch": 0.014871731317387532, "eval_loss": 3.0984115600585938, "eval_runtime": 7.2089, "eval_samples_per_second": 235.68, "eval_steps_per_second": 117.909, "step": 15 } ], "logging_steps": 1, "max_steps": 20, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 5, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 219472963043328.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }