© 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. 34-GEMMASCOPE-RES-131K
    4. 103629
    Prev
    Next
    INDEX
    Explanations

    supervised and unsupervised concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    清洁
    -0.77
     圣诞
    -0.72
    Presenta
    -0.71
    emp
    -0.70
    揺
    -0.69
    ریح
    -0.69
     Pedi
    -0.68
    SSI
    -0.68
    ective
    -0.68
     femeninos
    -0.66
    POSITIVE LOGITS
     supervised
    5.72
     supervision
    5.66
     supervise
    4.88
    supervised
    4.72
     Supervision
    4.50
     supervis
    4.34
     supervising
    4.34
    Supervision
    4.19
     supervisor
    3.73
     unsupervised
    3.66
    Activations Density 0.068%

    No Known Activations