© 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. 39840
    Prev
    Next
    INDEX
    Explanations

    .

    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
     take
    -0.61
     make
    -0.60
     use
    -0.58
     look
    -0.57
     check
    -0.56
     show
    -0.56
     read
    -0.54
     find
    -0.53
     choose
    -0.52
     create
    -0.51
    POSITIVE LOGITS
    æľ¬äºº
    0.30
     Spare
    0.25
    ä¸İæѤåIJĮæŶ
    0.25
    人士
    0.25
    orc
    0.25
     Strap
    0.24
    è°Ł
    0.24
    çĻ½çĻ½
    0.24
    ä¸įåħį
    0.24
    Õ¹
    0.24
    Activations Density 35.945%

    No Known Activations