© 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. 111781
    Prev
    Next
    INDEX
    Explanations

    Academic, technical documents

    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
    uyá»ģn
    -0.28
    cente
    -0.27
    uchar
    -0.26
     Libyan
    -0.24
    illas
    -0.23
    çļĩ
    -0.23
     agreeing
    -0.23
     boosts
    -0.23
    æĭį
    -0.23
    ]][
    -0.23
    POSITIVE LOGITS
    â
    0.27
    alto
    0.27
    wares
    0.27
     Klo
    0.25
     Pagination
    0.24
    åºĸ
    0.24
     Million
    0.24
    ä¸ĭéĿ¢å°±
    0.23
    æĢģåĬ¿
    0.23
    èĤĸ
    0.23
    Activations Density 0.000%

    No Known Activations