© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27556
    Prev
    Next
    INDEX
    Explanations

    row styles or content

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

    The neuron activates on “Row” (or “row”) tokens that are immediately followed by numeric identifiers, i.e. references to row numbers or row‐ID labels in code or data.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     There
    -3.36
    .
    -2.89
     هناك
    -2.83
    $
    -2.83
    U
    -2.56
    dzić
    -2.36
     还
    -2.30
    铼
    -2.23
     手機
    -2.19
     vorigen
    -2.11
    POSITIVE LOGITS
    妗
    2.55
     ит
    2.42
    ματα
    2.39
    
    2.36
    你来
    2.34
    ;
    2.34
    睺
    2.33
    樒
    2.30
    ‭
    2.30
    .":
    2.28
    Activations Density 0.020%

    No Known Activations