© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 66324
    Prev
    Next
    INDEX
    Explanations

    code variable interpolation

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

    This neuron activates on code‐style delimiters and syntax tokens—particularly braces (“{”, “{{”), parentheses (“(”), dollar signs (“$”), and similar structural symbols.

    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
    These
    -1.23
    Those
    -0.98
    還是
    -0.91
     этих
    -0.91
    itron
    -0.87
    
    -0.86
    ★★★
    -0.85
    ●●
    -0.85
     rheumat
    -0.85
    同様に
    -0.85
    POSITIVE LOGITS
     дори
    1.05
     Бор
    0.98
     of
    0.93
     Προ
    0.88
     their
    0.85
     also
    0.84
     бя
    0.84
     детский
    0.82
     diretor
    0.82
    Birth
    0.81
    Activations Density 0.010%

    No Known Activations