© 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. 89912
    Prev
    Next
    INDEX
    Explanations

    prepositions

    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
    inux
    -0.28
    (rr
    -0.28
    æ¿Ĥ
    -0.26
    路人
    -0.25
    rib
    -0.25
    æ¢Ń
    -0.25
    compass
    -0.24
    /ros
    -0.24
     PMC
    -0.24
    EVER
    -0.24
    POSITIVE LOGITS
    l
    0.32
     Cent
    0.27
    å¼Ĭ端
    0.26
    ioned
    0.26
    ãģĭãģĭãĤĭ
    0.26
    á»ĩ
    0.25
    _GAP
    0.25
    emi
    0.25
    dep
    0.25
    ä¸Ģåı¥è¯Ŀ
    0.25
    Activations Density 18.527%

    No Known Activations