© 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. 129233
    Prev
    Next
    INDEX
    Explanations

    The neuron detects words in “call-to-action” passages inviting reader feedback or engagement (e.g. let, know, comments, feedback, questions, appreciate).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    feedback or comments

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     ندار
    -1.06
    формы
    -1.05
     vulgares
    -1.03
     nuo
    -1.02
    gemeen
    -1.02
    golang
    -0.99
     }}">
    -0.99
    っていない
    -0.95
    IMIN
    -0.94
     برخی
    -0.94
    POSITIVE LOGITS
     feedback
    1.61
     any
    1.33
     comments
    1.22
     you
    1.17
     either
    1.16
     questions
    1.14
     appartamento
    1.13
     comment
    1.12
     Feedback
    1.12
    Feedback
    1.10
    Activations Density 0.031%

    No Known Activations