© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 78065
    Prev
    Next
    INDEX
    Explanations

    in case

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

    The neuron strongly activates on punctuation-like tokens (e.g. apostrophes and hyphens) and other short, formatting markers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    さえ
    -0.78
     συγκ
    -0.77
    INTA
    -0.72
     because
    -0.71
     issuer
    -0.71
     ufe
    -0.69
    persed
    -0.69
    тового
    -0.69
     Mancha
    -0.69
    igneurs
    -0.69
    POSITIVE LOGITS
     caso
    2.53
     case
    2.52
    case
    1.84
     Caso
    1.77
    Caso
    1.63
     apabila
    1.60
     случае
    1.54
    caso
    1.45
    Case
    1.45
     صورتی
    1.36
    Activations Density 0.015%

    No Known Activations