© 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. 22-GEMMASCOPE-RES-131K
    7. 27976
    Prev
    Next
    INDEX
    Explanations

    exit 0 or OK

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

    The neuron activates on numeric tokens (e.g. status codes, version numbers, line‐numbered values) appearing in code or annotations.

    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
     additional
    -1.34
     for
    -1.30
     to
    -1.09
     all
    -1.07
     from
    -1.05
    additional
    -1.05
     several
    -1.03
    balik
    -1.02
    haviours
    -1.00
    + 
    -0.97
    POSITIVE LOGITS
     OK
    1.41
    szám
    1.11
     okay
    1.11
    OK
    1.07
    お試し
    1.02
     Verd
    0.99
     évalu
    0.97
     fulfilment
    0.96
     favourable
    0.95
     ok
    0.95
    Activations Density 0.001%

    No Known Activations