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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 2562
    Prev
    Next
    INDEX
    Explanations

    code blocks and delimiters

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron consistently fires on the final token of each text segment (the closing punctuation or last word) – in other words, it marks the end-of-block token.

    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_24_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
    Introduction
    0.69
    Math
    0.68
    6
    0.68
    L
    0.68
    Six
    0.66
    Land
    0.66
    LT
    0.66
     sechs
    0.66
    Ground
    0.64
     पंतप्रधान
    0.64
    POSITIVE LOGITS
     यदि
    0.68
    ),
    0.68
    });
    0.68
    };
    0.67
     ።
    0.65
    ۔۔
    0.63
     }//
    0.63
    }};
    0.63
    },
    0.62
     noreferrer
    0.62
    Activations Density 0.001%

    No Known Activations