| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "eval_steps": 100, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.40908142924308777, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 1.4568, | |
| "mean_token_accuracy": 0.6849659357752119, | |
| "num_tokens": 535385.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.8686757683753967, | |
| "learning_rate": 0.0004877641290737884, | |
| "loss": 1.2429, | |
| "mean_token_accuracy": 0.7096964614731925, | |
| "num_tokens": 1070770.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.6107782125473022, | |
| "learning_rate": 0.00044700268840168044, | |
| "loss": 1.0227, | |
| "mean_token_accuracy": 0.7352448889187404, | |
| "num_tokens": 1606155.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.4725695550441742, | |
| "learning_rate": 0.00038247981605830125, | |
| "loss": 0.8815, | |
| "mean_token_accuracy": 0.7646138310432434, | |
| "num_tokens": 2141540.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.3993835747241974, | |
| "learning_rate": 0.0003019779227044398, | |
| "loss": 0.751, | |
| "mean_token_accuracy": 0.7935232503073556, | |
| "num_tokens": 2676925.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.3546556532382965, | |
| "learning_rate": 0.0002152067247599837, | |
| "loss": 0.6487, | |
| "mean_token_accuracy": 0.8254428488867623, | |
| "num_tokens": 3212310.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 0.3103465437889099, | |
| "learning_rate": 0.00013263210930352737, | |
| "loss": 0.5825, | |
| "mean_token_accuracy": 0.843472867352622, | |
| "num_tokens": 3747695.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.29546496272087097, | |
| "learning_rate": 6.421379363065141e-05, | |
| "loss": 0.5458, | |
| "mean_token_accuracy": 0.8587624430656433, | |
| "num_tokens": 4283080.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 0.2903711795806885, | |
| "learning_rate": 1.8204036358303172e-05, | |
| "loss": 0.5302, | |
| "mean_token_accuracy": 0.862883916922978, | |
| "num_tokens": 4818465.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.2829954922199249, | |
| "learning_rate": 1.5229324522605948e-07, | |
| "loss": 0.5272, | |
| "mean_token_accuracy": 0.8638145021029882, | |
| "num_tokens": 5353850.0, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.15744857767936e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |