© 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. 22-TRANSCODER-HP
    7. 130193
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çī¹çĤ¹
    -0.29
    åħ¼
    -0.27
     backups
    -0.26
    为主è¦ģ
    -0.26
    OPS
    -0.25
     ceil
    -0.24
    åıĬ
    -0.24
     Swipe
    -0.24
     such
    -0.24
    æµ¥
    -0.23
    POSITIVE LOGITS
    %p
    0.31
    .escape
    0.26
    escaped
    0.25
    .sap
    0.25
    ç»Ļä½łä»¬
    0.25
    Assignable
    0.24
    ÂĢÂĻ
    0.24
    arked
    0.24
    éĢĥ离
    0.24
    azines
    0.24
    Activations Density 0.002%

    No Known Activations