| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04655493482309125, | |
| "eval_steps": 4, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00186219739292365, | |
| "grad_norm": 11.199517250061035, | |
| "learning_rate": 1e-05, | |
| "loss": 14.6558, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00186219739292365, | |
| "eval_loss": 7.9340128898620605, | |
| "eval_runtime": 141.6164, | |
| "eval_samples_per_second": 3.199, | |
| "eval_steps_per_second": 0.402, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0037243947858473, | |
| "grad_norm": 11.964097023010254, | |
| "learning_rate": 2e-05, | |
| "loss": 17.5745, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00558659217877095, | |
| "grad_norm": 10.627275466918945, | |
| "learning_rate": 3e-05, | |
| "loss": 15.6694, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0074487895716946, | |
| "grad_norm": 12.379173278808594, | |
| "learning_rate": 4e-05, | |
| "loss": 16.864, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0074487895716946, | |
| "eval_loss": 7.823979377746582, | |
| "eval_runtime": 141.8419, | |
| "eval_samples_per_second": 3.194, | |
| "eval_steps_per_second": 0.402, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00931098696461825, | |
| "grad_norm": 12.158676147460938, | |
| "learning_rate": 5e-05, | |
| "loss": 13.9441, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0111731843575419, | |
| "grad_norm": 11.740424156188965, | |
| "learning_rate": 6e-05, | |
| "loss": 14.7925, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01303538175046555, | |
| "grad_norm": 12.046319007873535, | |
| "learning_rate": 7e-05, | |
| "loss": 14.4625, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0148975791433892, | |
| "grad_norm": 11.838757514953613, | |
| "learning_rate": 8e-05, | |
| "loss": 13.0003, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0148975791433892, | |
| "eval_loss": 6.556656837463379, | |
| "eval_runtime": 141.8338, | |
| "eval_samples_per_second": 3.194, | |
| "eval_steps_per_second": 0.402, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01675977653631285, | |
| "grad_norm": 15.124539375305176, | |
| "learning_rate": 9e-05, | |
| "loss": 14.9513, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0186219739292365, | |
| "grad_norm": 13.066884994506836, | |
| "learning_rate": 0.0001, | |
| "loss": 12.1102, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.020484171322160148, | |
| "grad_norm": 12.15008544921875, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 7.8592, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0223463687150838, | |
| "grad_norm": 9.749792098999023, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 7.4251, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0223463687150838, | |
| "eval_loss": 3.782747507095337, | |
| "eval_runtime": 141.8503, | |
| "eval_samples_per_second": 3.194, | |
| "eval_steps_per_second": 0.402, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.024208566108007448, | |
| "grad_norm": 12.579347610473633, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 8.1218, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0260707635009311, | |
| "grad_norm": 10.893141746520996, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 6.6247, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.027932960893854747, | |
| "grad_norm": 9.343505859375, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 4.8733, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0297951582867784, | |
| "grad_norm": 9.972923278808594, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 5.6742, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0297951582867784, | |
| "eval_loss": 2.0279412269592285, | |
| "eval_runtime": 141.8282, | |
| "eval_samples_per_second": 3.194, | |
| "eval_steps_per_second": 0.402, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03165735567970205, | |
| "grad_norm": 7.69146203994751, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 4.0756, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0335195530726257, | |
| "grad_norm": 6.7342753410339355, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 2.9383, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.035381750465549346, | |
| "grad_norm": 6.815323352813721, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 3.1479, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.037243947858473, | |
| "grad_norm": 8.356672286987305, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 3.6462, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.037243947858473, | |
| "eval_loss": 1.4177134037017822, | |
| "eval_runtime": 141.8183, | |
| "eval_samples_per_second": 3.194, | |
| "eval_steps_per_second": 0.402, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03910614525139665, | |
| "grad_norm": 8.037931442260742, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 3.1491, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.040968342644320296, | |
| "grad_norm": 5.409158229827881, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.4042, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.04283054003724395, | |
| "grad_norm": 5.937676906585693, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 2.8971, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0446927374301676, | |
| "grad_norm": 7.154982089996338, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 3.2611, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0446927374301676, | |
| "eval_loss": 1.2748059034347534, | |
| "eval_runtime": 141.8118, | |
| "eval_samples_per_second": 3.194, | |
| "eval_steps_per_second": 0.402, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04655493482309125, | |
| "grad_norm": 5.526090145111084, | |
| "learning_rate": 0.0, | |
| "loss": 2.0829, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.26732322177024e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |