© 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. 52967
    Prev
    Next
    INDEX
    Explanations

    Code and licenses

    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
    宪
    -0.26
    eec
    -0.26
     Character
    -0.26
    character
    -0.25
    #error
    -0.25
    .Character
    -0.25
    buffers
    -0.25
    icap
    -0.24
    enburg
    -0.24
     Bender
    -0.24
    POSITIVE LOGITS
    å¯ĵ
    0.33
    ÙĥØ©
    0.28
     rule
    0.28
    æİĮæı¡
    0.27
    ient
    0.27
     assert
    0.26
     imp
    0.26
    ç½ijç«Ļé¦ĸ页
    0.25
    å±Ģ
    0.25
     scientific
    0.25
    Activations Density 0.017%

    No Known Activations