© 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-3-4B
    5. 29-GEMMASCOPE-2-RES-16K
    7. 2578
    Prev
    Next
    INDEX
    Explanations

    knowing or understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-pt/resid_post/layer_29_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Instead
    0.81
     royale
    0.76
    代わりに
    0.75
    instead
    0.75
     instead
    0.75
    Instead
    0.74
     Overall
    0.73
    }></
    0.72
     unworthy
    0.72
    可以说
    0.70
    POSITIVE LOGITS
     instinctively
    1.95
     intimately
    1.93
     intuitively
    1.87
     firsthand
    1.64
     vaguely
    1.36
     intellectually
    1.28
     whereof
    1.25
     better
    1.24
     personally
    1.21
    ledged
    1.21
    Activations Density 0.304%

    No Known Activations