© 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. 22-GEMMASCOPE-RES-131K
    4. 71346
    Prev
    Next
    INDEX
    Explanations

    if IE, include, Condition, example, @ platform

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

    The neuron is spotting code‐level directive markers and comment or section‐header tokens (e.g. “@” in CSS/Sass/Swift, “//” comments, “[” INI headers, “ife” in makefiles, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Nä
    -1.18
     PWN
    -1.16
    Läs
    -1.16
    Tä
    -1.13
    griffe
    -1.08
    akyti
    -1.07
     kablo
    -1.07
    banez
    -1.06
    backers
    -1.05
     možnost
    -1.05
    POSITIVE LOGITS
    it
    1.11
     G
    1.03
     no
    1.02
     de
    1.02
     M
    1.01
     not
    1.01
    ´
    1.00
     only
    0.97
     esfuer
    0.96
    G
    0.96
    Activations Density 0.002%

    No Known Activations