© 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. 39735
    Prev
    Next
    INDEX
    Explanations

    in

    np_max-act-logits · gemini-2.0-flash
    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
    伤
    -0.46
    åıĹ伤
    -0.39
     injury
    -0.38
     injuries
    -0.37
    åĤ·
    -0.36
     injuring
    -0.35
     wound
    -0.35
    伤åı£
    -0.34
     injured
    -0.34
     hurt
    -0.33
    POSITIVE LOGITS
    inet
    0.29
    thest
    0.28
    career
    0.28
    kah
    0.27
    æī£
    0.27
    JOR
    0.26
    cate
    0.26
    itu
    0.26
    antu
    0.26
    ercul
    0.26
    Activations Density 0.013%

    No Known Activations