© 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. 53716
    Prev
    Next
    INDEX
    Explanations

    varied text snippets

    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
    次
    -0.26
    -Se
    -0.26
     vess
    -0.26
    oven
    -0.25
    沸èħ¾
    -0.25
    éĿ¢åŃĶ
    -0.25
    åĪĽå®¢
    -0.24
    æĹ¥ãģ«
    -0.24
    沸
    -0.24
    æĪIJè¯Ń
    -0.24
    POSITIVE LOGITS
     spare
    0.27
    éĹ²
    0.25
    ella
    0.25
     rac
    0.25
     rho
    0.24
    elda
    0.23
     Marshal
    0.23
    xBF
    0.23
    inka
    0.23
    å¾ĹäºĨ
    0.23
    Activations Density 0.005%

    No Known Activations