© 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. 128133
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on mentions of organized initiatives—especially the word “program” (and related terms like “initiative”) indicating a structured project or offering.

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

    programs and initiatives

    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
     család
    -1.16
     protože
    -1.13
    ówczas
    -1.12
     pełni
    -1.11
     large
    -1.05
    𝙤
    -1.03
    দ
    -1.02
     various
    -1.02
    흰색
    -1.02
     most
    -1.00
    POSITIVE LOGITS
     program
    1.78
     programs
    1.37
    program
    1.25
     programme
    1.23
     launched
    1.22
     Vuelta
    1.22
     whereby
    1.16
     scheme
    1.16
     initiative
    1.15
     scala
    1.13
    Activations Density 0.074%

    No Known Activations