© 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. 20-TRANSCODER-HP
    7. 107926
    Prev
    Next
    INDEX
    Explanations

    IDs and references

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     concepts
    -0.28
    åĩ¡æĺ¯
    -0.26
    èĻļæŀĦ
    -0.26
    ç»ĵåIJĪ
    -0.25
    åĪĨæĶ¯æľºæŀĦ
    -0.25
    -packed
    -0.24
    æĺ¯åIJ¦
    -0.24
     pad
    -0.24
    æĿ¥è®²
    -0.23
    禧
    -0.23
    POSITIVE LOGITS
     Booker
    0.32
    ercul
    0.30
    dns
    0.26
     BUG
    0.25
    NSS
    0.24
     Pole
    0.24
     Implemented
    0.24
     dns
    0.24
    æ·±åħ¥å¼Ģå±ķ
    0.23
    èĥĮ
    0.23
    Activations Density 12.068%

    No Known Activations