© 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. 32-TRANSCODER-HP
    7. 42424
    Prev
    Next
    INDEX
    Explanations

    non-alphanumeric characters

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æŃ£æĸĩ
    -0.26
    clude
    -0.25
    body
    -0.25
    caff
    -0.24
    ÑĤоп
    -0.24
    æĬĺ
    -0.23
    æĭĬ
    -0.23
     Reform
    -0.23
    umbo
    -0.23
    纵
    -0.23
    POSITIVE LOGITS
    羣åģĩ
    0.26
    æijĨæĶ¾
    0.24
    hyth
    0.24
    inel
    0.24
    ###↵
    0.24
    åľ¨é¦Ļ港
    0.23
    insky
    0.23
     symb
    0.23
     Robbins
    0.23
    çļĦéľĢæ±Ĥ
    0.23
    Activations Density 0.001%

    No Known Activations