© 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. 16324
    Prev
    Next
    INDEX
    Explanations

    abstract summary sections

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

    This neuron triggers on standalone numeric values (e.g. measurements, counts, page numbers, or parameter figures) appearing as document metadata or specification entries.

    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
     !.
    -0.94
    ter
    -0.86
     Metho
    -0.79
    (‘
    -0.79
    ting
    -0.77
    fa
    -0.77
    !,
    -0.77
     ?,
    -0.77
    RESULT
    -0.76
    ?;
    -0.76
    POSITIVE LOGITS
    Bedankt
    1.00
    時間の
    0.95
    iedene
    0.90
    echos
    0.88
     déploy
    0.84
    afy
    0.83
    latimes
    0.82
    ять
    0.82
    Mijn
    0.81
    entuk
    0.81
    Activations Density 0.004%

    No Known Activations