© 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. Qwen3-4B
    5. 35-TRANSCODER-HP
    7. 39834
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tracts
    -0.26
    ç»ĵåIJĪèµ·æĿ¥
    -0.24
    ourg
    -0.24
    æĥĭ
    -0.24
    Vectorizer
    -0.23
     combining
    -0.23
    èĮº
    -0.23
    ç»ĵåIJĪ
    -0.23
     âĨ
    -0.23
     disadvantage
    -0.23
    POSITIVE LOGITS
    éĺ³åİ¿
    0.28
    ASI
    0.26
    alion
    0.25
    gnore
    0.25
    ä¸ĢåĪ»
    0.25
     consistency
    0.24
     inconsist
    0.24
     valide
    0.24
    ä¸Ģ次æĢ§
    0.23
    ованиÑı
    0.23
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.