© 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. 34-GEMMASCOPE-RES-131K
    7. 82748
    Prev
    Next
    INDEX
    Explanations

    Candid, duplication, conjugates, -ates

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

    text containing "Candid" or forms of words with "lic" and "ate" appearing together, particularly in scientific and technical terminology.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    一会儿
    -0.72
    明顯
    -0.71
     hierdie
    -0.70
    QueryBuilder
    -0.67
     Topeka
    -0.67
     Departmental
    -0.66
    yped
    -0.66
    uchtigkeit
    -0.65
    bietern
    -0.65
    丢失
    -0.64
    POSITIVE LOGITS
    ates
    1.91
    ATES
    1.63
    ate
    1.61
    ATE
    1.33
    inate
    1.13
    licate
    1.04
    ciate
    1.00
    locate
    0.98
    bate
    0.93
    icate
    0.91
    Activations Density 0.043%

    No Known Activations