© 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. 22-TRANSCODER-HP
    7. 97884
    Prev
    Next
    INDEX
    Explanations

    Software licenses and code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oit
    -0.27
     Bosch
    -0.27
     Joined
    -0.25
    åIJĽ
    -0.25
     Junk
    -0.24
     каÑĩ
    -0.24
    <Pair
    -0.24
     bey
    -0.24
    yne
    -0.24
    æѤæĸĩ
    -0.23
    POSITIVE LOGITS
    tracts
    0.27
    ılan
    0.26
    ulously
    0.26
    ixa
    0.26
    нÑĥл
    0.25
     imper
    0.24
    è¿ĺ以为
    0.24
    æĺİçĻ½
    0.24
     facets
    0.24
    çŃīåĢĻ
    0.24
    Activations Density 0.013%

    No Known Activations