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. 74351
    Prev
    Next
    INDEX
    Explanations

    say "ant" words

    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
    liness
    -0.68
    INESS
    -0.65
    yship
    -0.64
     ALR
    -0.64
    RuleContext
    -0.63
    izations
    -0.63
    ties
    -0.63
    izers
    -0.62
    iness
    -0.61
    ization
    -0.61
    POSITIVE LOGITS
    ant
    1.19
    ANT
    0.99
    antly
    0.98
    anten
    0.80
    ants
    0.79
    antin
    0.72
    antic
    0.70
    ante
    0.70
    antes
    0.70
     ant
    0.69
    Activations Density 0.017%

    No Known Activations