© 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. 12438
    Prev
    Next
    INDEX
    Explanations

    see followed by words

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

    The neuron fires on instances of the verb “see” (especially in “I don’t see…”) where the writer expresses inability to understand or notice something.

    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
     to
    -1.66
     Nonetheless
    -1.55
    珖
    -1.52
    insegna
    -1.52
    ~•
    -1.49
    udahkan
    -1.48
     ثلاثة
    -1.45
     these
    -1.43
    VERTISEMENT
    -1.41
    這些
    -1.41
    POSITIVE LOGITS
    e
    1.89
    ang
    1.63
    or
    1.57
    ot
    1.57
    ic
    1.54
    ip
    1.52
     possibility
    1.52
    ap
    1.52
     którzy
    1.50
    1
    1.50
    Activations Density 0.010%

    No Known Activations