© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12163
    Prev
    Next
    INDEX
    Explanations

    gray, grey, grayscale

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

    The neuron fires on the hexadecimal digits that make up CSS‐style color codes.

    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
    密切
    0.59
     強
    0.58
    र्जी
    0.56
    氓
    0.55
     poisons
    0.54
     straighten
    0.53
     straightened
    0.53
    uent
    0.52
     ふ
    0.52
    逆
    0.52
    POSITIVE LOGITS
    ฤษ
    0.69
    email
    0.59
    rieb
    0.58
    emails
    0.56
    black
    0.55
    EMAIL
    0.55
    ここ
    0.55
    DateTime
    0.54
    date
    0.54
    datetime
    0.54
    Activations Density 0.007%

    No Known Activations