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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 20-CLT-HP
    4. 74431
    Prev
    Next
    INDEX
    Explanations

    _{

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AndEndTag
    -0.66
     artikkelen
    -0.65
    charging
    -0.64
    ItemLayout
    -0.61
     الرياضيه
    -0.61
     hospitalization
    -0.60
     Dö
    -0.60
    bleau
    -0.60
    casting
    -0.60
    Fiche
    -0.60
    POSITIVE LOGITS
     properly
    0.54
    expandindo
    0.47
     well
    0.44
    RegressionTest
    0.43
     Properly
    0.39
    properly
    0.39
     endwhile
    0.38
     adequately
    0.36
    loem
    0.36
    styleable
    0.36
    Activations Density 0.001%

    No Known Activations