© 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. 15777
    Prev
    Next
    INDEX
    Explanations

    legalese and technical terms

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

    The neuron fires on relatively rare or domain‐specific tokens (i.e. infrequent, content‐bearing words).

    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
    ↵↵↵
    -1.00
     distinctly
    -0.91
    влении
    -0.91
     когато
    -0.89
     პ
    -0.88
     tuyên
    -0.88
     while
    -0.87
    ությ
    -0.86
    richText
    -0.85
     vieles
    -0.84
    POSITIVE LOGITS
    ewon
    0.94
    ोड़
    0.87
     سنة
    0.87
     ſt
    0.86
    voke
    0.83
    υχ
    0.83
     YEAR
    0.82
     and
    0.82
    motions
    0.81
    hawaii
    0.80
    Activations Density 0.028%

    No Known Activations