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. 81
    Prev
    Next
    INDEX
    Explanations

    aster

    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
    ç§ijåѦ
    -0.29
    rip
    -0.29
    line
    -0.25
    ape
    -0.25
    åĺ¶
    -0.25
    æįĤ
    -0.25
    low
    -0.24
    etter
    -0.24
    istrict
    -0.23
    å¤ļåªĴä½ĵ
    -0.23
    POSITIVE LOGITS
     tran
    0.27
     Dich
    0.27
    DAL
    0.27
     tails
    0.27
     nfl
    0.24
    æĮĤ
    0.24
     tail
    0.24
    ICH
    0.23
    ,Yes
    0.23
    utherland
    0.23
    Activations Density 23.143%

    No Known Activations