© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 7907
    Prev
    Next
    INDEX
    Explanations

    disease of the

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

    phrases discussing "cancer of the" or "carcinoma of the" specific body parts.

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

    The neuron is primarily responsive to numeric tokens (quantities, measurements, percentages) 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_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
     a
    -1.73
     and
    -1.55
     this
    -1.45
     кілька
    -1.43
     which
    -1.42
     betek
    -1.41
     result
    -1.41
     includes
    -1.41
    طع
    -1.36
     créées
    -1.34
    POSITIVE LOGITS
     the
    1.70
     boneka
    1.55
    ⍽
    1.53
    selben
    1.51
     أثناء
    1.47
    therosclerosis
    1.47
    2
    1.46
    橚
    1.46
     of
    1.45
    不僅
    1.45
    Activations Density 0.179%

    No Known Activations