Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 20-GEMMASCOPE-TRANSCODER-16K
    4. 14170
    Prev
    Next
    INDEX
    Explanations

    the word "highly" and words near it, but not consistently

    oai_token-act-pair · gemini-2.0-flash

    highly

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_20/width_16k
    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
    حياتها
    -0.61
     désormais
    -0.57
    volmente
    -0.56
     découver
    -0.54
    eaways
    -0.54
    lıkla
    -0.54
    EndInit
    -0.53
     totally
    -0.53
     Super
    -0.52
     leçons
    -0.52
    POSITIVE LOGITS
    клопе
    0.72
     surla
    0.58
    تقاوى
    0.52
    DoubleQuotes
    0.52
     placebo
    0.52
    Filmografie
    0.52
     unpopular
    0.51
     sugges
    0.51
    dataclass
    0.51
     Teut
    0.50
    Activations Density 0.234%

    No Known Activations