© 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. 43266
    Prev
    Next
    INDEX
    Explanations

    thesis

    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
    åľº
    -0.30
    anic
    -0.28
    utherford
    -0.26
    habi
    -0.25
    imizer
    -0.24
    æľįåĬ¡æ°´å¹³
    -0.24
    åIJ¬è¯Ŀ
    -0.24
    /cs
    -0.24
    iability
    -0.23
    èı½
    -0.23
    POSITIVE LOGITS
     statement
    0.34
    声æĺİ
    0.29
    æĴ°
    0.28
    iola
    0.28
     concern
    0.27
    	statement
    0.27
    .syn
    0.27
    (statement
    0.27
     stating
    0.27
     paper
    0.27
    Activations Density 0.001%

    No Known Activations