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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33588
    Prev
    Next
    INDEX
    Explanations

    either or best

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

    The neuron activates on discourse markers that set up alternatives or hedges (e.g. “either,” “or,” “at best,” commas in those clauses).

    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
    ñales
    -0.98
     zwięks
    -0.98
    崁
    -0.96
     inilah
    -0.95
     but
    -0.92
     isoliert
    -0.92
    begin
    -0.91
     erity
    -0.89
    今後も
    -0.87
    ukh
    -0.86
    POSITIVE LOGITS
     outright
    1.59
     entweder
    1.32
    Either
    1.32
     downright
    1.31
     либо
    1.27
     either
    1.26
     Either
    1.16
    either
    1.09
     alternatively
    1.02
    要么
    1.02
    Activations Density 0.038%

    No Known Activations