© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 83888
    Prev
    Next
    INDEX
    Explanations

    data, age, peace, talk, store

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

    The neuron fires on any token carrying or adjacent to a numeric value (i.e. it detects numbers and number‐laden tokens).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     our
    -1.64
     those
    -1.56
     having
    -1.36
     possibly
    -1.35
     may
    -1.32
     strongly
    -1.31
    arschijnlijk
    -1.30
     their
    -1.30
     something
    -1.27
     Because
    -1.26
    POSITIVE LOGITS
    whom
    1.47
     Comentarios
    1.46
    ⛦
    1.45
     unicórnio
    1.45
     estadounid
    1.44
    urther
    1.41
     bessere
    1.41
    presumably
    1.38
     musia
    1.38
    experienced
    1.38
    Activations Density 0.047%

    No Known Activations