© 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. 33326
    Prev
    Next
    INDEX
    Explanations

    names connected by and

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

    The neuron detects numeric reference markers in academic citations (e.g. years, page or volume numbers) that follow author names.

    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
     at
    -3.02
     on
    -2.66
     as
    -2.63
     Despite
    -2.48
     from
    -2.47
     by
    -2.45
     or
    -2.44
     in
    -2.44
     both
    -2.42
     before
    -2.41
    POSITIVE LOGITS
     horrid
    2.33
     diriger
    2.22
     Dinas
    2.22
     alcune
    2.13
    饸
    2.09
     deliciously
    2.08
     faciliter
    2.06
     dando
    2.06
     ludicrous
    2.06
     alcuni
    2.03
    Activations Density 0.001%

    No Known Activations