| { |
| "MSE": 0.0, |
| "MSE/layer0": 0.0, |
| "dead_code_fraction": 1.0, |
| "dead_code_fraction/layer0": 1.0, |
| "epoch": 10.43, |
| "eval_MSE/layer0": 6073.8636798095695, |
| "eval_accuracy": 0.5791389432485323, |
| "eval_dead_code_fraction/layer0": 0.1981725, |
| "eval_input_norm/layer0": 0.7182212994247673, |
| "eval_loss": 1.6604058742523193, |
| "eval_multicode_k": 1, |
| "eval_output_norm/layer0": 76.78913438796998, |
| "eval_runtime": 6.7146, |
| "eval_samples": 100, |
| "eval_samples_per_second": 14.893, |
| "eval_steps_per_second": 0.298, |
| "input_norm": 0.0, |
| "input_norm/layer0": 0.0, |
| "loss": 1.6774777018260956, |
| "max_norm": 153.29054260253906, |
| "max_norm/layer0": 153.29054260253906, |
| "mean_norm": 75.17323780059814, |
| "mean_norm/layer0": 75.17323780059814, |
| "multicode_k": 1, |
| "output_norm": 0.0, |
| "output_norm/layer0": 0.0, |
| "perplexity": 5.261445896555633, |
| "runtime": 132212.7109, |
| "samples_per_second": 72.61, |
| "steps_per_second": 0.756, |
| "train_samples": 920563 |
| } |
