© 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. 10-GEMMASCOPE-RES-131K
    7. 21270
    Prev
    Next
    INDEX
    Explanations

    screenwriter and dehydrogenase contexts

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

    professional titles and technical terms, particularly words like "screenwriter" and "dehydrogenase".

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

    The neuron most strongly activates on the occupational term “screenwriter.”

    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
    Ꮄ
    -3.30
     camiset
    -2.80
    茀
    -2.75
    旰
    -2.70
    迨
    -2.67
     informar
    -2.67
    彞
    -2.67
     strategis
    -2.52
    橚
    -2.48
    婓
    -2.44
    POSITIVE LOGITS
    0
    2.78
     они
    2.53
    }$
    2.53
    in
    2.23
    釹
    2.23
    <tr>
    2.23
    ית
    2.23
     solche
    2.23
     $-$
    2.19
     那
    2.16
    Activations Density 0.013%

    No Known Activations