© 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. 111979
    Prev
    Next
    INDEX
    Explanations

    relationships

    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
    iable
    -0.27
    adio
    -0.26
     middle
    -0.26
     area
    -0.26
    è¾¾åĪ°äºĨ
    -0.25
     jint
    -0.24
     prominently
    -0.24
     lá»ĩ
    -0.24
    çĴ
    -0.23
     sola
    -0.23
    POSITIVE LOGITS
    ç»Ļèĩªå·±
    0.35
    çĽ¸ä¼´
    0.33
    好åıĭ
    0.33
    对èĩªå·±
    0.33
    åįıåIJĮ
    0.33
    çŁ¥å·±
    0.32
    ä¼Ļä¼´
    0.32
    éĻªä¼´
    0.32
    ç«ŀäºī对æīĭ
    0.31
    éĻªä½ł
    0.31
    Activations Density 0.016%

    No Known Activations