© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 120152
    Prev
    Next
    INDEX
    Explanations

    public, protected, static, void, function

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

    whitespace or line breaks preceding method or function definitions in code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron does not reliably activate on any tokens—it appears to be effectively “dead” and doesn’t detect a specific pattern.

    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
     kulinar
    -1.92
     katastro
    -1.92
     muelle
    -1.88
     Dernière
    -1.81
    蚨
    -1.77
     tunik
    -1.76
     kooper
    -1.75
     spektak
    -1.71
    archiw
    -1.70
    NÍ
    -1.70
    POSITIVE LOGITS
    k
    2.20
    5
    2.16
    3
    2.13
    8
    2.08
    1
    2.08
    us
    2.05
    6
    2.03
    l
    2.02
    on
    1.93
    p
    1.92
    Activations Density 0.005%

    No Known Activations