© 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. 70831
    Prev
    Next
    INDEX
    Explanations

    know what

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

    The neuron strongly responds to the sequence “you know all the/of…?”—i.e. rhetorical, second-person “you know all of” question phrasing.

    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
    வர்
    -1.27
     frequentemente
    -1.15
    "",
    -1.15
    鰾
    -1.15
     stille
    -1.10
    брать
    -1.09
    barkeit
    -1.08
     souhait
    -1.07
     thé
    -1.06
    nanoTime
    -1.06
    POSITIVE LOGITS
     but
    1.49
     will
    1.36
     reason
    1.32
     enough
    1.30
     including
    1.22
     achieved
    1.19
     variety
    1.16
    ULLY
    1.16
    chappe
    1.16
     that
    1.16
    Activations Density 0.056%

    No Known Activations