© 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. 22-GEMMASCOPE-RES-131K
    7. 75217
    Prev
    Next
    INDEX
    Explanations

    interrogative words and determine

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

    The neuron fires on words and phrases that signal determining or identifying something (e.g. identify, indication, figure out, determine).

    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
     ذلك
    -1.02
    DAFTAR
    -0.96
    cean
    -0.93
    furter
    -0.93
    罣
    -0.90
    それは
    -0.90
     hacerlo
    -0.88
    存于
    -0.87
    ivity
    -0.87
     farlo
    -0.86
    POSITIVE LOGITS
     how
    1.79
     what
    1.48
     why
    1.27
     your
    1.27
    ншот
    1.10
     possible
    1.09
     who
    1.07
    窺
    1.00
     vilka
    0.99
     where
    0.98
    Activations Density 0.031%

    No Known Activations