Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Qwen3-4B
    3. 1-TRANSCODER-HP
    4. 130
    Prev
    Next
    INDEX
    Explanations

    modeling

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_1.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.1.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çķĻ
    -0.27
    æĸĩä¸Ń
    -0.26
    isse
    -0.25
    oga
    -0.25
     nas
    -0.25
    çŁĽçĽ¾
    -0.25
    èµł
    -0.25
    çļĦæĦıæĢĿ
    -0.25
    ol
    -0.25
    çŰ
    -0.25
    POSITIVE LOGITS
    ebin
    0.27
    ="@
    0.26
    rapper
    0.26
     strm
    0.26
    andbox
    0.25
    )!↵
    0.25
    ordable
    0.25
    upd
    0.24
     pled
    0.24
    .spotify
    0.24
    Activations Density 27.103%

    No Known Activations