© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 15983
    Prev
    Next
    INDEX
    Explanations

    end of sentence punctuation

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

    The neuron activates on numeric tokens (especially multi-digit numbers or numbers with decimals).

    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
    CKS
    -1.18
     КА
    -1.17
    chose
    -1.16
    ﮯ
    -1.13
     médicaments
    -1.13
     Especial
    -1.13
     ALREADY
    -1.13
     disques
    -1.13
    jarati
    -1.12
    especial
    -1.11
    POSITIVE LOGITS
     this
    1.47
     that
    1.45
     at
    1.38
     other
    1.32
    </b>
    1.27
     these
    1.27
    That
    1.26
     if
    1.25
     only
    1.23
     all
    1.23
    Activations Density 0.003%

    No Known Activations