© 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. 107597
    Prev
    Next
    INDEX
    Explanations

    Ordinal numbers

    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
     GetMessage
    -0.26
    mits
    -0.25
    è°¯
    -0.25
    <IEnumerable
    -0.24
    Nib
    -0.24
    祯
    -0.24
    Ĥ¬
    -0.23
    æģ£
    -0.23
    nicos
    -0.23
    å°±åľ¨äºİ
    -0.23
    POSITIVE LOGITS
    èĭ¥æľī
    0.28
    åĪĨæĪIJ
    0.26
     half
    0.24
    agrid
    0.24
    Ùĥ
    0.24
     split
    0.24
    ir
    0.24
    伤
    0.23
     than
    0.23
     exact
    0.23
    Activations Density 8.591%

    No Known Activations