© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 5-CLT-HP
    7. 1941
    Prev
    Next
    INDEX
    Explanations

    Start of documents/articles

    np_max-act · 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
     undoubtedly
    -0.99
     undeniably
    -0.96
     unquestionably
    -0.88
     Undoubtedly
    -0.86
     undoubted
    -0.82
    certainly
    -0.79
     certainly
    -0.77
     TextAppearance
    -0.76
     undeniable
    -0.76
     zweifel
    -0.74
    POSITIVE LOGITS
     instead
    0.56
    ьаж
    0.50
    databind
    0.49
    ModelAdmin
    0.49
    ньому
    0.49
    Turns
    0.48
     invece
    0.47
     warning
    0.47
    刚好
    0.47
     basically
    0.47
    Activations Density 0.007%

    No Known Activations