© 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. 10-GEMMASCOPE-RES-131K
    4. 104191
    Prev
    Next
    INDEX
    Explanations

    code and technical discussion

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

    whitespace or blank lines, particularly in formatted documents like code, emails, or structured data.

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

    The neuron consistently fires on standalone multi‐digit numeric tokens (e.g. page or section 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
    2
    -4.19
    )
    -3.39
    8
    -3.06
    5
    -2.86
    7
    -2.83
    3
    -2.75
    4
    -2.73
    1
    -2.70
    </b>
    -2.63
    6
    -2.63
    POSITIVE LOGITS
    地道
    2.66
     anunciado
    2.31
     rigue
    2.30
     agricultores
    2.28
     achten
    2.28
     semblent
    2.27
     sabias
    2.27
     the
    2.25
     Gegensatz
    2.20
     Allgemeinen
    2.20
    Activations Density 0.099%

    No Known Activations