| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0616689150125265, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0030834457506263247, | |
| "grad_norm": 0.3808819055557251, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9657, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0030834457506263247, | |
| "eval_loss": 1.9325377941131592, | |
| "eval_runtime": 159.7296, | |
| "eval_samples_per_second": 3.425, | |
| "eval_steps_per_second": 1.715, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0061668915012526495, | |
| "grad_norm": 0.35360875725746155, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8817, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.009250337251878975, | |
| "grad_norm": 0.34825238585472107, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9287, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.012333783002505299, | |
| "grad_norm": 0.32264071702957153, | |
| "learning_rate": 4e-05, | |
| "loss": 1.8953, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.015417228753131625, | |
| "grad_norm": 0.32675665616989136, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9004, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015417228753131625, | |
| "eval_loss": 1.9066282510757446, | |
| "eval_runtime": 6.7924, | |
| "eval_samples_per_second": 80.531, | |
| "eval_steps_per_second": 40.339, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01850067450375795, | |
| "grad_norm": 0.35199978947639465, | |
| "learning_rate": 6e-05, | |
| "loss": 1.8666, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.021584120254384276, | |
| "grad_norm": 0.31535008549690247, | |
| "learning_rate": 7e-05, | |
| "loss": 1.8433, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.024667566005010598, | |
| "grad_norm": 0.30079585313796997, | |
| "learning_rate": 8e-05, | |
| "loss": 1.8836, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.027751011755636924, | |
| "grad_norm": 0.28127431869506836, | |
| "learning_rate": 9e-05, | |
| "loss": 1.8224, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03083445750626325, | |
| "grad_norm": 0.2556057870388031, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8895, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03083445750626325, | |
| "eval_loss": 1.8460979461669922, | |
| "eval_runtime": 6.806, | |
| "eval_samples_per_second": 80.371, | |
| "eval_steps_per_second": 40.259, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03391790325688957, | |
| "grad_norm": 0.22862684726715088, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.8543, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0370013490075159, | |
| "grad_norm": 0.24351704120635986, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.7939, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04008479475814222, | |
| "grad_norm": 0.24571087956428528, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.7667, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.04316824050876855, | |
| "grad_norm": 0.28732648491859436, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.8403, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.046251686259394874, | |
| "grad_norm": 0.2647162675857544, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7692, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.046251686259394874, | |
| "eval_loss": 1.8046153783798218, | |
| "eval_runtime": 6.903, | |
| "eval_samples_per_second": 79.241, | |
| "eval_steps_per_second": 39.693, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.049335132010021196, | |
| "grad_norm": 0.2547331154346466, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.8054, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.052418577760647525, | |
| "grad_norm": 0.29585474729537964, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.7471, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.05550202351127385, | |
| "grad_norm": 0.2694046199321747, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.8154, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.058585469261900176, | |
| "grad_norm": 0.2696322798728943, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.7992, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0616689150125265, | |
| "grad_norm": 0.24875527620315552, | |
| "learning_rate": 0.0, | |
| "loss": 1.8184, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0616689150125265, | |
| "eval_loss": 1.7931327819824219, | |
| "eval_runtime": 6.7128, | |
| "eval_samples_per_second": 81.486, | |
| "eval_steps_per_second": 40.818, | |
| "step": 20 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2999758891253760.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |