© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 267
    Prev
    Next
    INDEX
    Explanations

    This neuron detects added‐line markers in unified diffs—i.e. lines beginning with “+” (often including added code like “+ echo …”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    胛
    0.44
     、,
    0.42
    物质
    0.41
    ळख
    0.41
    êmes
    0.41
    ẖ
    0.41
     amplo
    0.41
    コハマ
    0.40
     prêts
    0.40
    र्यंत
    0.40
    POSITIVE LOGITS
    }
    1.20
    };
    1.03
    },
    0.97
    })
    0.95
    });
    0.91
    }.
    0.87
    }//
    0.80
    }}
    0.79
    //}
    0.78
    }\
    0.75
    Activations Density 0.002%

    No Known Activations