© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 130866
    Prev
    Next
    INDEX
    Explanations

    medical, technical and cosmetic terms

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

    technical terms in programming and interface code contexts.

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

    The neuron fires most strongly on unusually long single tokens (e.g. long mixed-case identifiers like “OnItemSelectedListener” or long words like “iridescent” or “AccessControl”).

    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
    </td>
    -2.61
     και
    -2.52
    —
    -2.44
     the
    -2.28
    n
    -2.25
    In
    -2.23
    2
    -2.20
     *
    -2.11
     и
    -2.09
     Taking
    -2.03
    POSITIVE LOGITS
    媗
    2.61
    粜
    2.45
    蔸
    2.39
    颏
    2.38
     breathtaking
    2.38
    嚒
    2.36
     dismal
    2.33
    匐
    2.33
     différents
    2.28
     unending
    2.27
    Activations Density 0.000%

    No Known Activations