© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 9697
    Prev
    Next
    INDEX
    Explanations

    coding and medical contexts

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

    scientific and technical notation in specialized fields like chemistry, biology, and programming.

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

    The neuron selectively activates on tokens that are or contain numerical values (e.g. digits, decimals, counts).

    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
     recargable
    -0.96
    </>
    
    -0.93
    물
    -0.90
    /***
    
    -0.89
    бур
    -0.88
    erste
    -0.88
    德
    -0.87
    setcounter
    -0.86
     kuu
    -0.85
    poietic
    -0.85
    POSITIVE LOGITS
     Leurs
    1.02
    它们的
    0.90
     Their
    0.86
     dame
    0.85
     Leur
    0.85
    拎
    0.85
    SIGMA
    0.84
     authors
    0.84
     gitara
    0.83
    尔夫
    0.82
    Activations Density 0.001%

    No Known Activations