© 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. 129847
    Prev
    Next
    INDEX
    Explanations

    Conversational text excerpts

    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
    èĢĮæĺ¯
    -0.70
    åıªæĺ¯
    -0.48
     instead
    -0.40
     Instead
    -0.40
    åıªæĺ¯ä¸Ģ个
    -0.40
     merely
    -0.40
    Instead
    -0.40
     juste
    -0.39
     purely
    -0.39
    åįķ纯
    -0.37
    POSITIVE LOGITS
    è¿ĺæĺ¯è¦ģ
    0.30
    è¿ĺä¸įæĺ¯
    0.30
    åĬĽè¿ĺæĺ¯
    0.27
    --){↵
    0.26
     Enlight
    0.26
    æĿ¥ä¸įåıĬ
    0.26
    ä¸įåĸľæ¬¢
    0.25
    tml
    0.25
    è¿ĺ没æľī
    0.24
     anyway
    0.24
    Activations Density 0.002%

    No Known Activations