© 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. 48342
    Prev
    Next
    INDEX
    Explanations

    deal with

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

    The neuron fires most strongly on longer, specialized jargon terms (e.g. technical methods, function names, chemical‐style words, obscure proper nouns) rather than common everyday words.

    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
     but
    -1.65
     there
    -1.52
     then
    -1.38
    says
    -1.37
    knows
    -1.34
    founded
    -1.33
    studied
    -1.30
    helped
    -1.30
    emphas
    -1.30
    expect
    -1.28
    POSITIVE LOGITS
     would
    1.48
     Along
    1.48
     börja
    1.43
     and
    1.41
     также
    1.40
     neuen
    1.36
     tillsammans
    1.34
     Also
    1.33
     Because
    1.32
     Alongside
    1.30
    Activations Density 0.196%

    No Known Activations