© 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. 80670
    Prev
    Next
    INDEX
    Explanations

    photo attribution or source

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

    The neuron detects image‐credit markers and photo attribution tokens (e.g. “Photo,” “Getty,” “Source,” and numeric credit values).

    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
     “
    -1.41
    ുന്നു
    -1.38
    をはじめ
    -1.36
    蟶
    -1.35
    还想
    -1.35
    想到了
    -1.33
    }));
    -1.32
    ㇼ
    -1.29
    をご覧ください
    -1.27
    品です
    -1.27
    POSITIVE LOGITS
    )
    1.40
     jų
    1.22
    évalu
    1.21
     protease
    1.20
    Viitteet
    1.20
    ธิ
    1.17
    tokenId
    1.16
    (
    1.16
    pering
    1.16
    iforn
    1.14
    Activations Density 0.023%

    No Known Activations