© 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. 34-GEMMASCOPE-RES-131K
    7. 58907
    Prev
    Next
    INDEX
    Explanations

    starts or ends with punctuation

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

    The neuron fires on numeric constants and parameter‐assignment syntax—i.e. numbers (often with units), equals signs, parentheses, and related tokens in model or code definitions.

    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
     so
    -1.02
     what
    -1.02
     first
    -0.92
     encourage
    -0.90
     genuinely
    -0.83
     even
    -0.82
     encourages
    -0.81
     one
    -0.81
     bring
    -0.79
     truly
    -0.78
    POSITIVE LOGITS
     plafon
    0.98
    влению
    0.93
     triom
    0.91
     urat
    0.90
     suppos
    0.90
     DEEP
    0.90
    Verwendung
    0.89
    grazie
    0.88
    ckets
    0.88
     immen
    0.88
    Activations Density 0.005%

    No Known Activations