© 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. 63259
    Prev
    Next
    INDEX
    Explanations

    period

    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
    YM
    -0.26
    IVA
    -0.26
    æŀķ头
    -0.26
    éĥ¨åī¯
    -0.25
    APA
    -0.25
    appa
    -0.24
    è§Ħ模æľĢ大
    -0.24
    ?",↵
    -0.24
    èĦļä¸ĭ
    -0.24
    è¿ĻæĿ¡
    -0.24
    POSITIVE LOGITS
    æĸ°åŁºå»º
    0.24
    ṡ
    0.24
    izzo
    0.24
    esse
    0.24
    reactstrap
    0.24
     Keeps
    0.23
     gou
    0.23
    comp
    0.23
    otics
    0.22
    çĿĽ
    0.22
    Activations Density 80.858%

    No Known Activations