© 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. 10-GEMMASCOPE-RES-131K
    7. 103681
    Prev
    Next
    INDEX
    Explanations

    leech, fstream, horseradish, SSN neurons

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

    The neuron detects specific scientific and natural words, particularly "leech," "horseradish," "fstream," and programming syntax elements like "Parser".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    t
    -3.13
    0
    -3.09
    4
    -2.52
    otropy
    -2.45
     放題
    -2.38
     merito
    -2.38
     précis
    -2.38
    7
    -2.36
    他的
    -2.17
     romana
    -2.16
    POSITIVE LOGITS
     They
    3.06
     From
    2.69
     тези
    2.64
    臜
    2.64
     It
    2.58
    phors
    2.58
     allerdings
    2.55
    罴
    2.52
    $,
    2.45
     имеет
    2.36
    Activations Density 0.008%

    No Known Activations