© 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. 111623
    Prev
    Next
    INDEX
    Explanations

    center

    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
    že
    -0.26
     sve
    -0.25
    ório
    -0.24
    åĬ¨æīĭ
    -0.24
    .Toast
    -0.23
    sea
    -0.23
    ática
    -0.23
    SEA
    -0.23
    /no
    -0.23
    enced
    -0.22
    POSITIVE LOGITS
    ey
    0.29
     Jehovah
    0.26
    çľ¼è§Ĵ
    0.26
    æĮĤåľ¨
    0.26
    å¦Ĭå¨ł
    0.25
    è¶´åľ¨
    0.25
     PKK
    0.25
    arshal
    0.25
     repay
    0.25
     Mund
    0.24
    Activations Density 0.021%

    No Known Activations