| { | |
| "d_sae": 4096, | |
| "actual_l0": 79.9205251, | |
| "effective_l0": 42, | |
| "cross_entropy": 228.38722282714843, | |
| "sae_type": "saebench", | |
| "sae_release": "adamkarvonen/saebench_gemma-2-2b_width-2pow12_date-0108", | |
| "num_tokens": 10000000, | |
| "num_epochs": 1, | |
| "lr": 0.0005, | |
| "loaded_dag": null, | |
| "allow_negative_influence": false, | |
| "history": { | |
| "train_loss": [ | |
| 373.0598709106445, | |
| 296.3568134852818, | |
| 271.22952662876673, | |
| 257.95599496023993, | |
| 249.27438975742885, | |
| 242.92079282488143, | |
| 238.3831261771066, | |
| 235.0788822719029, | |
| 232.04134761265345, | |
| 229.43816942487445, | |
| 229.34045748029436 | |
| ], | |
| "train_steps": [ | |
| 140, | |
| 280, | |
| 420, | |
| 560, | |
| 700, | |
| 840, | |
| 980, | |
| 1120, | |
| 1260, | |
| 1400, | |
| 1407 | |
| ], | |
| "val_loss": [ | |
| 317.3101373046875, | |
| 280.5291108398437, | |
| 263.3136259277344, | |
| 252.9333521484375, | |
| 245.83410717773438, | |
| 240.5973893798828, | |
| 236.54287941894532, | |
| 233.30607668457031, | |
| 230.65738371582032, | |
| 228.45383054199218, | |
| 228.38722282714843 | |
| ], | |
| "val_cross_entropy": [ | |
| 317.3101373046875, | |
| 280.5291108398437, | |
| 263.3136259277344, | |
| 252.9333521484375, | |
| 245.83410717773438, | |
| 240.5973893798828, | |
| 236.54287941894532, | |
| 233.30607668457031, | |
| 230.65738371582032, | |
| 228.45383054199218, | |
| 228.38722282714843 | |
| ], | |
| "effective_l0": [ | |
| 62, | |
| 54, | |
| 50, | |
| 47, | |
| 46, | |
| 44, | |
| 44, | |
| 43, | |
| 42, | |
| 42, | |
| 42 | |
| ], | |
| "val_steps": [ | |
| 140, | |
| 280, | |
| 420, | |
| 560, | |
| 700, | |
| 840, | |
| 980, | |
| 1120, | |
| 1260, | |
| 1400, | |
| 1407 | |
| ] | |
| }, | |
| "sae_id": "MatryoshkaBatchTopK_gemma-2-2b__0108/resid_post_layer_12/trainer_2" | |
| } |