| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01360451669954425, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000272090333990885, | |
| "eval_loss": 10.805745124816895, | |
| "eval_runtime": 96.9862, | |
| "eval_samples_per_second": 15.961, | |
| "eval_steps_per_second": 7.981, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013604516699544248, | |
| "grad_norm": 4.270561695098877, | |
| "learning_rate": 5e-05, | |
| "loss": 10.8453, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0027209033399088495, | |
| "grad_norm": 5.673886299133301, | |
| "learning_rate": 0.0001, | |
| "loss": 10.5568, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0027209033399088495, | |
| "eval_loss": 9.172842025756836, | |
| "eval_runtime": 97.5356, | |
| "eval_samples_per_second": 15.871, | |
| "eval_steps_per_second": 7.936, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004081355009863274, | |
| "grad_norm": 7.321413040161133, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 7.9529, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005441806679817699, | |
| "grad_norm": 5.264472007751465, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 4.4401, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.005441806679817699, | |
| "eval_loss": 2.9730587005615234, | |
| "eval_runtime": 98.6634, | |
| "eval_samples_per_second": 15.69, | |
| "eval_steps_per_second": 7.845, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006802258349772125, | |
| "grad_norm": 3.589726686477661, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.4232, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.008162710019726549, | |
| "grad_norm": 4.526918411254883, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0587, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.008162710019726549, | |
| "eval_loss": 0.8501867651939392, | |
| "eval_runtime": 97.6538, | |
| "eval_samples_per_second": 15.852, | |
| "eval_steps_per_second": 7.926, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009523161689680974, | |
| "grad_norm": 1.8098701238632202, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.5922, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.010883613359635398, | |
| "grad_norm": 3.461613178253174, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.6072, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.010883613359635398, | |
| "eval_loss": 0.44126075506210327, | |
| "eval_runtime": 97.8308, | |
| "eval_samples_per_second": 15.823, | |
| "eval_steps_per_second": 7.912, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.012244065029589824, | |
| "grad_norm": 2.419137716293335, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.4135, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01360451669954425, | |
| "grad_norm": 4.879242897033691, | |
| "learning_rate": 0.0, | |
| "loss": 0.7161, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01360451669954425, | |
| "eval_loss": 0.40443843603134155, | |
| "eval_runtime": 97.577, | |
| "eval_samples_per_second": 15.864, | |
| "eval_steps_per_second": 7.932, | |
| "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": 9080277024374784.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |