© 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. 112189
    Prev
    Next
    INDEX
    Explanations

    original

    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
    목
    -0.27
    éļ¨æĻĤ
    -0.25
    istor
    -0.25
    mits
    -0.25
    ajs
    -0.24
     sidelines
    -0.24
    .OUT
    -0.24
    neck
    -0.24
    ä¸ŃæľŁ
    -0.24
     Voc
    -0.24
    POSITIVE LOGITS
    iginal
    0.29
    ity
    0.28
     tally
    0.27
    轸
    0.26
     challenging
    0.26
    iners
    0.26
    庾
    0.25
    mente
    0.24
    稳妥
    0.24
    posit
    0.24
    Activations Density 0.175%

    No Known Activations