© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35501
    Prev
    Next
    INDEX
    Explanations

    people saying something

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

    This neuron flags substantive content words in main text or block-quoted passages (versus navigation, headers, or metadata).

    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
     epitom
    -2.00
     using
    -1.97
     on
    -1.88
     first
    -1.85
     in
    -1.83
     an
    -1.82
     Despite
    -1.80
     with
    -1.73
     from
    -1.72
     before
    -1.71
    POSITIVE LOGITS
     .........
    1.91
     .......
    1.90
    鵙
    1.88
     dando
    1.84
     joyous
    1.82
     ..........
    1.80
     Dinas
    1.78
     ridiculously
    1.75
     arrogant
    1.71
     wondrous
    1.70
    Activations Density 0.142%

    No Known Activations