© 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. 72175
    Prev
    Next
    INDEX
    Explanations

    code identifiers

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

    The neuron strongly activates on short, uppercase or alphanumeric identifier tokens (e.g. two‐ or three‐letter acronyms, variable/component names, version tags) in code.

    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
    昻
    -2.67
    未完
    -2.45
    ꯸
    -2.34
     всѣ
    -2.33
    圌
    -2.27
     química
    -2.25
    的吗
    -2.23
     моя
    -2.23
    藹
    -2.23
    䊐
    -2.22
    POSITIVE LOGITS
    逦
    2.44
     They
    2.42
    .
    2.38
    杀手
    2.09
     emphasis
    2.08
     Gedanke
    2.08
     Nachmittag
    2.06
    ly
    2.06
    st
    2.02
     Except
    2.00
    Activations Density 0.265%

    No Known Activations