© 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. 32-TRANSCODER-HP
    7. 42722
    Prev
    Next
    INDEX
    Explanations

    authors and their work

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¨Ģ
    -0.30
    åĨĻä½ľ
    -0.26
    æİĸ
    -0.26
    .syntax
    -0.26
    mn
    -0.25
    .Writer
    -0.24
    æı®
    -0.24
    asting
    -0.24
     Joi
    -0.24
    jb
    -0.24
    POSITIVE LOGITS
    è¿IJ
    0.28
    lei
    0.27
    edium
    0.27
    seys
    0.26
    edd
    0.25
    irl
    0.24
     alg
    0.24
    ç²¾åįİ
    0.24
     softball
    0.24
    enz
    0.24
    Activations Density 0.002%

    No Known Activations