© 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. 10-GEMMASCOPE-RES-131K
    7. 45216
    Prev
    Next
    INDEX
    Explanations

    disease steel from Introduction

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

    contractions and possessives with apostrophes (like don't, 's).

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron detects contractions—words ending with apostrophe-plus-letter sequences like ’re or n't.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     serisi
    -1.76
     geïsole
    -1.61
     encendido
    -1.61
    štění
    -1.60
     gesloten
    -1.58
    䛗
    -1.55
     espectáculos
    -1.55
    ⵔ
    -1.55
     SPAIN
    -1.55
    speakers
    -1.55
    POSITIVE LOGITS
    “
    2.06
     of
    2.05
     a
    1.95
    at
    1.85
     from
    1.79
     Their
    1.77
    ’
    1.70
    ”
    1.68
     this
    1.63
     Want
    1.62
    Activations Density 0.000%

    No Known Activations