© 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. 10-GEMMASCOPE-RES-131K
    7. 72656
    Prev
    Next
    INDEX
    Explanations

    species, years, and classifications

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

    years in scientific taxonomic references, particularly between author names and dates.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron detects numeric tokens (numbers and number‐related punctuation), i.e. occurrences of digits or numeric data 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_10/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
     palio
    -1.83
    棡
    -1.80
    
    -1.77
     kuda
    -1.76
     diza
    -1.75
     jaya
    -1.74
     diametro
    -1.73
    鹇
    -1.73
     stiker
    -1.72
    ㈲
    -1.70
    POSITIVE LOGITS
     during
    1.80
     but
    1.77
     Since
    1.67
     Usually
    1.67
     that
    1.67
     to
    1.66
     without
    1.63
     Being
    1.61
     However
    1.58
     at
    1.48
    Activations Density 0.040%

    No Known Activations