© 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. 39348
    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
    shots
    -0.26
    -Smith
    -0.26
     laat
    -0.25
    ãģ®ãģĵãģ¨
    -0.25
    ToSend
    -0.24
    utherland
    -0.24
    .TXT
    -0.24
    isp
    -0.24
    ensi
    -0.23
     Elections
    -0.23
    POSITIVE LOGITS
    Aux
    0.25
     proportions
    0.25
    大家åĪĨ享
    0.25
    è´¨éĩı
    0.25
     {};↵↵
    0.24
    _nd
    0.24
    éĩijéĴ±
    0.24
     wom
    0.24
    åĨĢ
    0.23
     mole
    0.23
    Activations Density 11.034%

    No Known Activations

    This feature has no known activations.