© 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. 97562
    Prev
    Next
    INDEX
    Explanations

    Describes origin, style, perspective

    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
    ired
    -0.25
    ãģĵãģ¨ãģ«
    -0.24
    äºĮ代
    -0.24
    [${
    -0.23
    [...,
    -0.23
    åįİ侨
    -0.23
    elsen
    -0.23
     Phạm
    -0.23
    URN
    -0.23
    alus
    -0.23
    POSITIVE LOGITS
     Segment
    0.26
     Seg
    0.25
     Ring
    0.25
    èĻº
    0.24
    catch
    0.23
     catch
    0.23
    çĸ¹
    0.23
    çļĦåĬªåĬĽ
    0.23
     segreg
    0.23
    eworld
    0.23
    Activations Density 0.050%

    No Known Activations