© 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. 129741
    Prev
    Next
    INDEX
    Explanations

    phone numbers or 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
    代表
    -0.28
    resents
    -0.27
     Represents
    -0.25
    æĸ¥
    -0.25
    æľ¬äºº
    -0.25
    wrapped
    -0.24
     endors
    -0.24
     represents
    -0.24
    bris
    -0.23
    è¹Ī
    -0.23
    POSITIVE LOGITS
     vac
    0.28
    vac
    0.27
    visitor
    0.26
    âĢĴ
    0.25
    grounds
    0.25
    xmm
    0.24
    å°½æĥħ
    0.24
    èģļ
    0.23
    å¹Į
    0.23
     Middle
    0.23
    Activations Density 0.027%

    No Known Activations