© 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. 64472
    Prev
    Next
    INDEX
    Explanations

    research papers and code

    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
    icho
    -0.28
     vuel
    -0.27
    ogenerated
    -0.24
     enough
    -0.24
    å¸ľ
    -0.24
     mount
    -0.23
    daq
    -0.23
    bill
    -0.23
    loo
    -0.23
    åĽº
    -0.23
    POSITIVE LOGITS
    ç²īç¢İ
    0.31
    æŶä¸į
    0.27
    åıªä¸įè¿ĩæĺ¯
    0.26
    ores
    0.26
    æŶéĻIJ
    0.25
    гаÑĢ
    0.25
     Holt
    0.24
    iat
    0.24
    ÐĶÐķ
    0.24
    каÑĢ
    0.24
    Activations Density 0.826%

    No Known Activations