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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21978
    Prev
    Next
    INDEX
    Explanations

    reportedly allegedly deemed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
    sei
    -0.06
    நà¯įதà¯ģ
    -0.06
    èĦļ
    -0.05
    UREMENT
    -0.05
    æĶľ
    -0.05
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    -0.05
    respect
    -0.05
    ئÙħ
    -0.05
    ående
    -0.05
     irresist
    -0.05
    POSITIVE LOGITS
    æį®
    0.09
    被认为
    0.09
    æĵļ
    0.08
     supposedly
    0.07
     reportedly
    0.07
    æį®è¯´
    0.07
    被认为æĺ¯
    0.07
     deemed
    0.06
     allegedly
    0.06
     seems
    0.06
    Activations Density 0.328%

    No Known Activations