© 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. 99674
    Prev
    Next
    INDEX
    Explanations

    distribution which shares, markets which have grown

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

    The neuron responds to clause-connecting tokens—especially commas, periods, and coordinating words like “so,” “and,” or “with”—marking boundaries between phrases.

    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
     including
    -1.79
     have
    -1.58
     in
    -1.56
     regarding
    -1.56
     as
    -1.48
     his
    -1.48
     having
    -1.48
     help
    -1.46
     a
    -1.44
     and
    -1.43
    POSITIVE LOGITS
    明日は
    1.52
    rió
    1.45
    今度
    1.42
    majority
    1.39
    knows
    1.38
    down
    1.38
     attirer
    1.35
     entirety
    1.34
     refuser
    1.33
     tirer
    1.30
    Activations Density 0.197%

    No Known Activations