© 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. 34-GEMMASCOPE-RES-131K
    7. 78275
    Prev
    Next
    INDEX
    Explanations

    words and code snippets

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

    The neuron is picking out programming‐code snippets (e.g. function calls, regex patterns, identifiers in code) rather than plain prose.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     翻
    -0.91
     关于
    -0.90
     werkzaam
    -0.82
    🐜
    -0.82
    dienne
    -0.79
     TimeUnit
    -0.79
     facilitator
    -0.79
    noty
    -0.79
    ppes
    -0.78
     avión
    -0.78
    POSITIVE LOGITS
     Janet
    0.76
     even
    0.76
    жать
    0.73
     especially
    0.73
     Margaret
    0.73
    VCR
    0.71
    眷
    0.71
    蚕
    0.71
    Agora
    0.71
    even
    0.71
    Activations Density 0.000%

    No Known Activations