© 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. 64307
    Prev
    Next
    INDEX
    Explanations

    wide

    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
     Skyl
    -0.26
    HEN
    -0.24
     Scal
    -0.24
     Care
    -0.24
    ALER
    -0.24
    æĥ¯ä¾ĭ
    -0.23
     Naturally
    -0.23
    æĥħ趣
    -0.23
     Exterior
    -0.23
     oltre
    -0.23
    POSITIVE LOGITS
    ÑĥÑĪки
    0.28
    riend
    0.28
    textbox
    0.26
    oha
    0.26
    gent
    0.26
    filled
    0.26
     strength
    0.25
    ä¸ĢæľŁ
    0.24
    -par
    0.24
    (range
    0.24
    Activations Density 16.802%

    No Known Activations