© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103338
    Prev
    Next
    INDEX
    Explanations

    software names and technical terms

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

    product codes, technical identifiers, and abbreviated names in structured data formats.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron is triggered by three-letter, all-caps substrings (e.g. “HIP,” “GET,” “MCN,” etc.).

    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
     that
    -1.61
     lamentable
    -1.31
     virš
    -1.28
     dieną
    -1.23
    rechte
    -1.23
     蓋
    -1.21
     вигля
    -1.21
    歺
    -1.20
     of
    -1.20
     ۔
    -1.20
    POSITIVE LOGITS
     also
    2.20
     également
    1.75
     también
    1.63
     on
    1.39
     ALSO
    1.38
     TAMBÉM
    1.37
     software
    1.34
     a
    1.33
     também
    1.33
     myös
    1.31
    Activations Density 0.069%

    No Known Activations