| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 0.4859062731266022, | |
| "learning_rate": 9.4375e-05, | |
| "loss": 2.4443, | |
| "mean_token_accuracy": 0.5860039107501507, | |
| "num_tokens": 40960.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 0.3083125948905945, | |
| "learning_rate": 8.8125e-05, | |
| "loss": 0.0556, | |
| "mean_token_accuracy": 0.9846604824066162, | |
| "num_tokens": 81920.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 0.24934740364551544, | |
| "learning_rate": 8.1875e-05, | |
| "loss": 0.0563, | |
| "mean_token_accuracy": 0.9847581863403321, | |
| "num_tokens": 122880.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.3064737915992737, | |
| "learning_rate": 7.5625e-05, | |
| "loss": 0.0497, | |
| "mean_token_accuracy": 0.9860527634620666, | |
| "num_tokens": 163840.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 0.26368093490600586, | |
| "learning_rate": 6.9375e-05, | |
| "loss": 0.0541, | |
| "mean_token_accuracy": 0.9845627784729004, | |
| "num_tokens": 204800.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.26650479435920715, | |
| "learning_rate": 6.3125e-05, | |
| "loss": 0.0448, | |
| "mean_token_accuracy": 0.9867611169815064, | |
| "num_tokens": 245760.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 0.2567497491836548, | |
| "learning_rate": 5.6875e-05, | |
| "loss": 0.0488, | |
| "mean_token_accuracy": 0.9860039114952087, | |
| "num_tokens": 286720.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.2256651371717453, | |
| "learning_rate": 5.0625e-05, | |
| "loss": 0.0488, | |
| "mean_token_accuracy": 0.9857840776443482, | |
| "num_tokens": 327680.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 0.26728323101997375, | |
| "learning_rate": 4.4375e-05, | |
| "loss": 0.0468, | |
| "mean_token_accuracy": 0.9865657091140747, | |
| "num_tokens": 368640.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "eval_runtime": 16.058, | |
| "eval_samples_per_second": 4.982, | |
| "eval_steps_per_second": 0.311, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.2592334449291229, | |
| "learning_rate": 3.8125e-05, | |
| "loss": 0.0475, | |
| "mean_token_accuracy": 0.9867366909980774, | |
| "num_tokens": 409600.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 0.2610887289047241, | |
| "learning_rate": 3.1875e-05, | |
| "loss": 0.0475, | |
| "mean_token_accuracy": 0.9867855429649353, | |
| "num_tokens": 450560.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.2812255322933197, | |
| "learning_rate": 2.5625e-05, | |
| "loss": 0.044, | |
| "mean_token_accuracy": 0.9877381563186646, | |
| "num_tokens": 491520.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 0.2330714762210846, | |
| "learning_rate": 1.9375e-05, | |
| "loss": 0.0457, | |
| "mean_token_accuracy": 0.9867122650146485, | |
| "num_tokens": 532480.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 0.2395644336938858, | |
| "learning_rate": 1.3125e-05, | |
| "loss": 0.0517, | |
| "mean_token_accuracy": 0.9855642437934875, | |
| "num_tokens": 573440.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.2232440710067749, | |
| "learning_rate": 6.875000000000001e-06, | |
| "loss": 0.0442, | |
| "mean_token_accuracy": 0.9875916004180908, | |
| "num_tokens": 614400.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.2189246416091919, | |
| "learning_rate": 6.25e-07, | |
| "loss": 0.0419, | |
| "mean_token_accuracy": 0.9886174917221069, | |
| "num_tokens": 655360.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_runtime": 15.1106, | |
| "eval_samples_per_second": 5.294, | |
| "eval_steps_per_second": 0.331, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 160, | |
| "total_flos": 0.0, | |
| "train_loss": 0.19822477344423534, | |
| "train_runtime": 202.9526, | |
| "train_samples_per_second": 1.577, | |
| "train_steps_per_second": 0.788 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |