© 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. 69136
    Prev
    Next
    INDEX
    Explanations

    what is/was/will

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

    The neuron strongly activates on the token “what” (and related copula tokens like “is/was”), i.e. it detects clauses introduced by “what” that set up explanatory or definitional statements.

    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
    何より
    -1.77
     広場
    -1.73
    begin
    -1.71
     hvilke
    -1.67
    
    -1.63
     ()
    -1.61
    ++
    -1.60
    ⸰
    -1.60
    これだけ
    -1.60
     it
    -1.60
    POSITIVE LOGITS
     “
    2.31
     —
    2.08
     "
    1.92
     a
    1.86
     //
    1.69
     «
    1.58
     That
    1.57
     an
    1.55
     –
    1.55
    ')
    1.54
    Activations Density 0.031%

    No Known Activations