© 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. 26689
    Prev
    Next
    INDEX
    Explanations

    policy and drug regulation

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

    technical documentation, formal articles, or professional content.

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

    The neuron fires most strongly on domain-specific, informational content words – technical jargon, specialized nouns and verbs in formal or how-to/academic write-ups (e.g. “exposing,” “complies,” “construction,” “Drug,” “tyres”).

    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
     with
    -1.11
     at
    -1.00
     also
    -0.93
     incorporating
    -0.90
     including
    -0.88
     on
    -0.87
     or
    -0.86
     in
    -0.85
     incorporate
    -0.84
    ūrė
    -0.84
    POSITIVE LOGITS
     BECAUSE
    1.11
     เพราะ
    0.96
    emph
    0.94
    看似
    0.93
     CONSIDER
    0.91
     DIDN
    0.91
     USED
    0.90
     ymm
    0.90
     why
    0.88
     echter
    0.88
    Activations Density 0.115%

    No Known Activations