© 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. 460
    Prev
    Next
    INDEX
    Explanations

    or reference or national

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

    The neuron fires on standalone numeric tokens—especially large multi‐digit numbers (e.g. version numbers or count values) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the word "and" in software license boilerplate text, particularly in the phrase "Redistribution and use in source and binary forms".

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    open-source software license boilerplate, especially the standard “redistribution and use… with or without modification” conditions found in license headers.

    oai_token-act-pair · gpt-5Triggered 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
     ransomware
    -1.38
     reptiles
    -1.37
     Rajasthan
    -1.32
     rheumatoid
    -1.28
     Reuters
    -1.26
     র
    -1.25
     rowspan
    -1.25
     rugby
    -1.23
     rumah
    -1.22
     reddish
    -1.22
    POSITIVE LOGITS
    read
    0.88
    rr
    0.88
    esserung
    0.86
    lhs
    0.85
    饋
    0.85
    RO
    0.81
    ro
    0.81
    ﺩ
    0.79
     snowfall
    0.79
    rec
    0.77
    Activations Density 0.322%

    No Known Activations