© 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. 107781
    Prev
    Next
    INDEX
    Explanations

    URLs and code

    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
     impulse
    -0.30
    uç
    -0.29
     rearr
    -0.28
    mı
    -0.27
    Merc
    -0.27
     analysis
    -0.27
     impulses
    -0.26
    pring
    -0.26
    beck
    -0.25
    ung
    -0.24
    POSITIVE LOGITS
    :].
    0.28
    åĪļ
    0.26
    ç¾İåĽ½äºº
    0.25
    (ev
    0.25
    ...',↵
    0.24
    ä¸įè¿ĩ
    0.24
    _SI
    0.24
    :↵↵↵↵↵↵
    0.23
    åī¯å±Ģéķ¿
    0.23
    头åıij
    0.23
    Activations Density 12.269%

    No Known Activations