© 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. 10-GEMMASCOPE-RES-131K
    7. 42172
    Prev
    Next
    INDEX
    Explanations

    autistic, drug, powered, functional

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

    words with the negative prefix "non-" or "un-" (and their derivatives like "unassisted" or "unsupporting").

    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_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
     rather
    -1.44
    Instead
    -1.37
    如果
    -1.30
    仳
    -1.27
    不管
    -1.27
    ariot
    -1.21
     boneco
    -1.20
     rewarding
    -1.20
            
    -1.19
    และ
    -1.18
    POSITIVE LOGITS
    的区别
    1.38
    んの
    1.31
     cherchés
    1.30
    ことはない
    1.29
     of
    1.25
    わけです
    1.24
     pracę
    1.22
     sés
    1.22
    berspace
    1.19
     imprimible
    1.17
    Activations Density 0.066%

    No Known Activations