© 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. 20-TRANSCODER-HP
    7. 64412
    Prev
    Next
    INDEX
    Explanations

    Mixed web data

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åħīæ»ij
    -0.27
    åıįè¿ĩæĿ¥
    -0.24
    -rounded
    -0.24
     shorthand
    -0.24
     Shorts
    -0.24
    ä»Ģä¹Īéĥ½ä¸į
    -0.24
     odby
    -0.24
    åıĶåıĶ
    -0.24
    aur
    -0.24
    fram
    -0.23
    POSITIVE LOGITS
    cient
    0.29
    erte
    0.27
    tical
    0.26
    BILE
    0.26
     RCS
    0.25
    ï¼ĮãĢĬ
    0.25
    çŃ¹
    0.24
     Rolex
    0.24
    _imm
    0.24
    ckt
    0.24
    Activations Density 1.446%

    No Known Activations