© 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. 108062
    Prev
    Next
    INDEX
    Explanations

    Article or blog post descriptions

    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
    illo
    -0.34
    bos
    -0.28
    oulos
    -0.26
    ıc
    -0.25
    ils
    -0.25
    bit
    -0.25
     run
    -0.25
    ilet
    -0.25
     foreground
    -0.25
    BITS
    -0.24
    POSITIVE LOGITS
    è¿Ļå¼ł
    0.26
    rams
    0.26
    having
    0.25
    ìĿ¸
    0.24
    null
    0.24
    缩
    0.24
    ì¤ijìķĻ
    0.23
    篾
    0.23
    wan
    0.23
    è¿Ļä¹Łæĺ¯
    0.23
    Activations Density 15.164%

    No Known Activations