© 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. 34364
    Prev
    Next
    INDEX
    Explanations

    listing numbers and descriptions

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

    the neuron detects multi‐digit numeric tokens in the text.

    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
    
    -2.75
     Dinas
    -2.55
     acheter
    -2.53
     italiens
    -2.50
    姈
    -2.42
     perioden
    -2.41
     faciliter
    -2.41
     WIL
    -2.39
    ఝ
    -2.36
    -:
    -2.33
    POSITIVE LOGITS
     or
    3.72
     and
    3.58
     from
    3.53
     with
    3.47
     which
    3.34
     in
    3.02
     through
    3.00
     at
    2.98
     before
    2.91
     without
    2.91
    Activations Density 0.066%

    No Known Activations