© 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. 80831
    Prev
    Next
    INDEX
    Explanations

    conductance | Wolfgang Pauli | Hana | laryngectomy

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

    proper names (like Wolfgang, Hana) and technical terms (conductance) in academic and biographical texts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on tokens that start with an uppercase letter (proper names, headings, or other capitalized terms).

    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
     dikutip
    -2.63
    猡
    -2.63
     pembeli
    -2.63
    塒
    -2.61
    Ꮵ
    -2.59
     When
    -2.58
     myſelf
    -2.47
    
    -2.47
     prossima
    -2.45
    鱚
    -2.44
    POSITIVE LOGITS
    ↵↵
    3.66
    .
    3.05
     -
    2.63
     wee
    2.59
     той
    2.58
    𖣔
    2.55
     immense
    2.55
     o
    2.42
     omission
    2.41
     of
    2.39
    Activations Density 0.007%

    No Known Activations