© 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. 22-GEMMASCOPE-RES-131K
    7. 34056
    Prev
    Next
    INDEX
    Explanations

    substitute decision, progenitor cell, general anesthesia

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

    medical and technical terms, particularly in scientific papers.

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

    The neuron activates on numeric tokens and number-like expressions in the text.

    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
    yeezy
    -0.89
     électriques
    -0.84
     understatement
    -0.83
    溅
    -0.82
    onomy
    -0.81
    lightweight
    -0.80
    Escola
    -0.79
    僮
    -0.79
     préstamos
    -0.78
    hova
    -0.77
    POSITIVE LOGITS
     if
    1.90
     necessary
    1.44
     because
    1.37
     backup
    1.22
     substitute
    1.20
     necessity
    1.20
     after
    1.20
     If
    1.17
     needed
    1.15
     fallback
    1.13
    Activations Density 0.174%

    No Known Activations