© 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. 22-GEMMASCOPE-RES-131K
    7. 129293
    Prev
    Next
    INDEX
    Explanations

    The neuron is sensitive to tokens containing digits (i.e. standalone numbers or version numbers) in the text.

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

    WITHOUT license terms

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    éu
    -0.90
    サイズは
    -0.88
     melap
    -0.87
     restre
    -0.86
    ester
    -0.86
    zere
    -0.86
     Voici
    -0.86
     flott
    -0.85
     indispensables
    -0.84
    lastName
    -0.84
    POSITIVE LOGITS
     spørs
    0.95
     grüne
    0.85
    ụp
    0.85
     dripping
    0.84
    crochet
    0.84
    Ngoài
    0.82
    elry
    0.82
    訃
    0.82
    outube
    0.82
     веке
    0.81
    Activations Density 0.002%

    No Known Activations