© 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. 10-GEMMASCOPE-RES-131K
    7. 14595
    Prev
    Next
    INDEX
    Explanations

    package fr and essay writing

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

    The neuron activates on isolated numeric literal tokens (e.g. the standalone “1”) in code or data contexts.

    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
     achieving
    -1.70
     through
    -1.65
     During
    -1.55
     attaining
    -1.52
    榀
    -1.44
    達は
    -1.41
     each
    -1.41
    ↬
    -1.40
     despite
    -1.39
     providing
    -1.38
    POSITIVE LOGITS
    ౩
    2.00
    
    1.80
     która
    1.77
    ་་
    1.75
    ఝ
    1.73
    糴
    1.72
    “
    1.72
    敺
    1.71
     który
    1.70
    涷
    1.68
    Activations Density 0.002%

    No Known Activations