© 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. 107138
    Prev
    Next
    INDEX
    Explanations

    History and geography of places

    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
    mans
    -0.28
    èĢħãģĮ
    -0.27
     tasked
    -0.26
    ìĪĺê°Ģ
    -0.26
    ä¹Ļçĥ¯
    -0.26
    comp
    -0.26
    ogne
    -0.25
    .FromResult
    -0.25
    客人
    -0.24
    nable
    -0.23
    POSITIVE LOGITS
    åį³ä½¿æĺ¯
    0.25
     even
    0.25
    è¿Ļä¸ĢçĤ¹
    0.24
    ]=-
    0.24
    .ResponseWriter
    0.24
    _hdl
    0.23
     synonym
    0.23
    (serial
    0.23
    æ°ij主
    0.23
    å°±è¿ŀ
    0.23
    Activations Density 16.739%

    No Known Activations