© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7754
    Prev
    Next
    INDEX
    Explanations

    formatting or past performance

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

    The neuron responds to content‐bearing (i.e. low‐frequency or “informative”) words—especially longer or more specific nouns and verbs—while ignoring common function words.

    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
    grep
    0.52
    df
    0.49
    поте
    0.48
    viable
    0.48
    (:
    0.47
    eligible
    0.46
    gos
    0.46
     liable
    0.46
    🙋
    0.46
     grep
    0.46
    POSITIVE LOGITS
    ORGE
    0.64
     Huh
    0.60
     Ibar
    0.59
    ަން
    0.57
     Međutim
    0.57
     Ezra
    0.57
     Cependant
    0.56
     羽
    0.56
    INUS
    0.56
    ="../../../../
    0.55
    Activations Density 0.258%

    No Known Activations