© 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. 33-TRANSCODER-HP
    7. 54297
    Prev
    Next
    INDEX
    Explanations

    antioxidants and chemical compounds

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¿¡
    -0.28
    /body
    -0.26
    ç»ı纪人
    -0.26
    <num
    -0.24
    ç»ı纪
    -0.24
    enums
    -0.24
     studs
    -0.24
    åĽ¾æĸĩ
    -0.23
     Flake
    -0.23
    >window
    -0.23
    POSITIVE LOGITS
     power
    0.35
     res
    0.33
     ep
    0.29
     processes
    0.27
     bif
    0.27
    让èĩªå·±
    0.27
    ase
    0.26
    power
    0.25
     arab
    0.25
     Ep
    0.25
    Activations Density 0.021%

    No Known Activations