© 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. 35-TRANSCODER-HP
    7. 39930
    Prev
    Next
    INDEX
    Explanations

    Code/Unusual Characters

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    vos
    -0.28
    izable
    -0.25
    æ¡£
    -0.25
     yö
    -0.25
    ARA
    -0.24
    æī¿
    -0.23
    cade
    -0.23
    OMET
    -0.23
    ow
    -0.23
    Articles
    -0.23
    POSITIVE LOGITS
    edBy
    0.30
    appiness
    0.28
    åıªçŁ¥éģĵ
    0.28
    æį¨ãģ¦
    0.28
    agina
    0.27
    agers
    0.27
    æįŁ
    0.27
    amm
    0.26
    ä¸ĩåĪĨ
    0.26
    ä¸įå°ij人
    0.25
    Activations Density 0.241%

    No Known Activations