© 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. 34-GEMMASCOPE-RES-131K
    7. 63049
    Prev
    Next
    INDEX
    Explanations

    numbers and punctuation preceding lists

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

    The neuron is tracking how far into the document a token appears (i.e. its absolute character/position offset).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     of
    -1.88
     in
    -1.87
     on
    -1.73
    ,
    -1.66
     the
    -1.66
     all
    -1.54
     as
    -1.51
     just
    -1.48
     The
    -1.46
     a
    -1.45
    POSITIVE LOGITS
     étan
    1.98
     refroid
    1.75
    ↵↵
    1.73
    Inhalts
    1.65
     précip
    1.64
     cuarentena
    1.60
     étanche
    1.59
    Datenschutz
    1.59
    Kommentare
    1.55
     manquer
    1.55
    Activations Density 0.047%

    No Known Activations