© 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. 142057
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     пÑģ
    -0.29
    edic
    -0.26
     Entire
    -0.25
     polo
    -0.25
    ],&
    -0.25
    unist
    -0.24
    éĴĿ
    -0.24
    éĶIJ
    -0.24
    ÑģÑĤÑĥп
    -0.24
     hend
    -0.23
    POSITIVE LOGITS
     midnight
    0.28
    å«Į
    0.28
    ulum
    0.26
    ighbors
    0.25
    umber
    0.24
    atism
    0.24
    armor
    0.24
    ittel
    0.24
    ignment
    0.24
    éĻįä½İæĪIJæľ¬
    0.23
    Activations Density 6.359%

    No Known Activations

    This feature has no known activations.