© 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. 18-TRANSCODER-HP
    7. 89831
    Prev
    Next
    INDEX
    Explanations

    Distance

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.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»Ĩ
    -0.26
    doch
    -0.26
    åĪĨå·¥
    -0.25
    çļĦåľ°æĸ¹
    -0.25
    perms
    -0.25
    éĤ¶
    -0.25
    æ¥Ķ
    -0.24
    继ç»Ń
    -0.24
    EH
    -0.24
    .datatables
    -0.24
    POSITIVE LOGITS
    ivism
    0.28
     according
    0.27
    baugh
    0.27
    missive
    0.27
    semble
    0.27
    æıIJ
    0.24
     modern
    0.24
    æ²½
    0.24
    çݰ代åĨľä¸ļ
    0.24
    æĶ¥
    0.24
    Activations Density 6.769%

    No Known Activations