{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.08837825894829872, "eval_steps": 10, "global_step": 50, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0017675651789659744, "eval_loss": 1.5736385583877563, "eval_runtime": 18.5113, "eval_samples_per_second": 12.911, "eval_steps_per_second": 6.483, "step": 1 }, { "epoch": 0.008837825894829872, "grad_norm": 1.6883810758590698, "learning_rate": 5e-05, "loss": 1.4734, "step": 5 }, { "epoch": 0.017675651789659744, "grad_norm": 1.110305666923523, "learning_rate": 0.0001, "loss": 1.4599, "step": 10 }, { "epoch": 0.017675651789659744, "eval_loss": 1.436456561088562, "eval_runtime": 18.9204, "eval_samples_per_second": 12.632, "eval_steps_per_second": 6.342, "step": 10 }, { "epoch": 0.026513477684489615, "grad_norm": 1.0780854225158691, "learning_rate": 9.619397662556435e-05, "loss": 1.3273, "step": 15 }, { "epoch": 0.03535130357931949, "grad_norm": 1.3645310401916504, "learning_rate": 8.535533905932738e-05, "loss": 1.2836, "step": 20 }, { "epoch": 0.03535130357931949, "eval_loss": 1.2428770065307617, "eval_runtime": 18.9823, "eval_samples_per_second": 12.591, "eval_steps_per_second": 6.322, "step": 20 }, { "epoch": 0.04418912947414936, "grad_norm": 0.8436744213104248, "learning_rate": 6.91341716182545e-05, "loss": 1.2168, "step": 25 }, { "epoch": 0.05302695536897923, "grad_norm": 0.8301315307617188, "learning_rate": 5e-05, "loss": 1.2328, "step": 30 }, { "epoch": 0.05302695536897923, "eval_loss": 1.2236849069595337, "eval_runtime": 18.9713, "eval_samples_per_second": 12.598, "eval_steps_per_second": 6.325, "step": 30 }, { "epoch": 0.0618647812638091, "grad_norm": 0.879909336566925, "learning_rate": 3.086582838174551e-05, "loss": 1.2017, "step": 35 }, { "epoch": 0.07070260715863898, "grad_norm": 1.6775568723678589, "learning_rate": 1.4644660940672627e-05, "loss": 1.1638, "step": 40 }, { "epoch": 0.07070260715863898, "eval_loss": 1.2111330032348633, "eval_runtime": 19.0169, "eval_samples_per_second": 12.568, "eval_steps_per_second": 6.31, "step": 40 }, { "epoch": 0.07954043305346885, "grad_norm": 0.7706012725830078, "learning_rate": 3.8060233744356633e-06, "loss": 1.0841, "step": 45 }, { "epoch": 0.08837825894829872, "grad_norm": 0.9315032362937927, "learning_rate": 0.0, "loss": 1.2157, "step": 50 }, { "epoch": 0.08837825894829872, "eval_loss": 1.2092931270599365, "eval_runtime": 18.9727, "eval_samples_per_second": 12.597, "eval_steps_per_second": 6.325, "step": 50 } ], "logging_steps": 5, "max_steps": 50, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 13, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.1051146382671872e+16, "train_batch_size": 2, "trial_name": null, "trial_params": null }