© 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. 30556
    Prev
    Next
    INDEX
    Explanations

    All rights reserved

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

    The neuron predominantly activates on numeric tokens (e.g. version numbers, years, IDs), so it is detecting sequences of digits.

    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
    மான
    -0.82
     حل
    -0.69
     Sections
    -0.69
     SECTIONS
    -0.68
    ਾਰ
    -0.68
     löytyy
    -0.67
    Sections
    -0.67
    scribe
    -0.67
    ான
    -0.65
     Basal
    -0.65
    POSITIVE LOGITS
    ありがとうございます
    0.75
     futbolista
    0.70
    hér
    0.70
     jails
    0.70
    中等
    0.69
    шения
    0.68
     renounced
    0.67
    FAIR
    0.67
    uninstall
    0.66
     універ
    0.66
    Activations Density 0.050%

    No Known Activations