© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38294
    Prev
    Next
    INDEX
    Explanations

    number of occurrences

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

    The neuron fires on technical terms around counting or locating occurrences (e.g. “occurrences,” “subsequence,” “counts,” “number of”).

    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
     to
    -1.07
    谑
    -0.96
     leurs
    -0.91
     and
    -0.87
    完成です
    -0.86
    emos
    -0.85
    ruppo
    -0.84
    Ὀ
    -0.84
    Urine
    -0.84
     their
    -0.83
    POSITIVE LOGITS
     counting
    1.63
     count
    1.43
    count
    1.41
    COUNT
    1.35
     counted
    1.33
    Count
    1.33
     Count
    1.23
     COUNT
    1.23
    有多少
    1.20
     counts
    1.16
    Activations Density 0.029%

    No Known Activations