© 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. 141519
    Prev
    Next
    INDEX
    Explanations

    Relative clauses introducers

    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
    ãģ¶ãĤĬ
    -0.29
    èĨ
    -0.27
    ľ
    -0.26
    ç®ĹäºĨ
    -0.25
    ä½ľèĢħæīĢæľī
    -0.24
     Yuri
    -0.24
    											↵
    -0.24
    						č↵
    -0.24
    ÑĤол
    -0.24
    Modal
    -0.23
    POSITIVE LOGITS
    ç§ijæķĻ
    0.28
    -aos
    0.27
    agnost
    0.26
    assed
    0.26
    æĤ¬æµ®
    0.26
     Rolling
    0.25
    åĮĸåIJĪ
    0.25
     barrier
    0.25
     Barrier
    0.24
     oft
    0.23
    Activations Density 14.259%

    No Known Activations