© 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. 129771
    Prev
    Next
    INDEX
    Explanations

    Code or astronomy

    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
    ateg
    -0.27
    ylv
    -0.26
    elligent
    -0.26
    atisfied
    -0.26
    è´
    -0.25
    achable
    -0.25
    .fit
    -0.24
    _beh
    -0.24
     Station
    -0.24
     bothered
    -0.23
    POSITIVE LOGITS
    onis
    0.28
    åŁİå¸Ĥ
    0.26
    å¸Ĥå§Ķ书记
    0.26
    æĿĥ
    0.25
    穹
    0.25
    ä¸įæĪIJ
    0.25
    inja
    0.24
    æĥ¯ä¾ĭ
    0.24
    æľ«
    0.23
    endo
    0.23
    Activations Density 0.010%

    No Known Activations