© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 119008
    Prev
    Next
    INDEX
    Explanations

    parsing tokens

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

    the word "tokens" in programming code and technical documentation.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates on numeric tokens (digits or numbers).

    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
    o
    -3.03
    ed
    -2.89
    as
    -2.67
    g
    -2.34
    ,”
    -2.33
    嫱
    -2.30
    ie
    -2.28
    el
    -2.27
    ti
    -2.27
    </
    -2.22
    POSITIVE LOGITS
     asesor
    2.78
    薷
    2.64
     SEÑ
    2.55
    茀
    2.52
     COMERC
    2.44
     MÁS
    2.41
     imperative
    2.38
    写真は
    2.33
    s
    2.31
    了
    2.30
    Activations Density 0.005%

    No Known Activations