© 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. 129989
    Prev
    Next
    INDEX
    Explanations

    Requests/Inquiries

    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
    çļĦåľ°æĸ¹
    -0.28
    SEP
    -0.27
    Clip
    -0.26
    Õ¹
    -0.24
     Heading
    -0.24
    :numel
    -0.24
     Spare
    -0.24
     Ain
    -0.24
    åĨĽäºº
    -0.24
    sass
    -0.23
    POSITIVE LOGITS
    åıĪ被
    0.31
    交
    0.31
    çķ´
    0.26
    已被
    0.25
    æıIJåĩº
    0.25
    ivid
    0.25
    交æĺĵæīĢ
    0.25
    æıIJæ¡Ī
    0.24
    å·¥ä½ľç»Ħ
    0.24
    å·²ç»ı被
    0.24
    Activations Density 6.953%

    No Known Activations