{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.10741138560687433, "eval_steps": 10, "global_step": 50, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0021482277121374865, "eval_loss": 9.146766662597656, "eval_runtime": 6.1234, "eval_samples_per_second": 32.008, "eval_steps_per_second": 16.004, "step": 1 }, { "epoch": 0.010741138560687433, "grad_norm": 48.73690414428711, "learning_rate": 5e-05, "loss": 9.132, "step": 5 }, { "epoch": 0.021482277121374866, "grad_norm": 42.14037322998047, "learning_rate": 0.0001, "loss": 7.8659, "step": 10 }, { "epoch": 0.021482277121374866, "eval_loss": 6.615852355957031, "eval_runtime": 6.0593, "eval_samples_per_second": 32.347, "eval_steps_per_second": 16.173, "step": 10 }, { "epoch": 0.0322234156820623, "grad_norm": 37.33824157714844, "learning_rate": 9.619397662556435e-05, "loss": 5.2623, "step": 15 }, { "epoch": 0.04296455424274973, "grad_norm": 27.734710693359375, "learning_rate": 8.535533905932738e-05, "loss": 3.4119, "step": 20 }, { "epoch": 0.04296455424274973, "eval_loss": 2.691011905670166, "eval_runtime": 6.0969, "eval_samples_per_second": 32.148, "eval_steps_per_second": 16.074, "step": 20 }, { "epoch": 0.05370569280343716, "grad_norm": 34.26387023925781, "learning_rate": 6.91341716182545e-05, "loss": 2.3496, "step": 25 }, { "epoch": 0.0644468313641246, "grad_norm": 14.376835823059082, "learning_rate": 5e-05, "loss": 1.7896, "step": 30 }, { "epoch": 0.0644468313641246, "eval_loss": 1.6869527101516724, "eval_runtime": 6.1944, "eval_samples_per_second": 31.641, "eval_steps_per_second": 15.821, "step": 30 }, { "epoch": 0.07518796992481203, "grad_norm": 9.836889266967773, "learning_rate": 3.086582838174551e-05, "loss": 1.3319, "step": 35 }, { "epoch": 0.08592910848549946, "grad_norm": 9.571033477783203, "learning_rate": 1.4644660940672627e-05, "loss": 1.9569, "step": 40 }, { "epoch": 0.08592910848549946, "eval_loss": 1.3932305574417114, "eval_runtime": 6.2087, "eval_samples_per_second": 31.569, "eval_steps_per_second": 15.784, "step": 40 }, { "epoch": 0.0966702470461869, "grad_norm": 15.935542106628418, "learning_rate": 3.8060233744356633e-06, "loss": 1.4733, "step": 45 }, { "epoch": 0.10741138560687433, "grad_norm": 10.49211311340332, "learning_rate": 0.0, "loss": 1.3353, "step": 50 }, { "epoch": 0.10741138560687433, "eval_loss": 1.3013057708740234, "eval_runtime": 6.126, "eval_samples_per_second": 31.995, "eval_steps_per_second": 15.997, "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": 1278959891251200.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }