© 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. 30-TRANSCODER-HP
    7. 96562
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     AUTHORS
    -0.30
     dow
    -0.26
    noinspection
    -0.26
     Pot
    -0.25
    pon
    -0.25
    é¼¹
    -0.25
    åIJł
    -0.25
    atest
    -0.25
     Hayward
    -0.25
    è¯Ŀ说
    -0.24
    POSITIVE LOGITS
    åĽ´ç»ķ
    0.34
    åĮħåĽ´
    0.32
    ç§ĥ
    0.27
     surround
    0.27
    æĵĬ
    0.27
    _FP
    0.26
    åı£è¢ĭ
    0.25
     пок
    0.25
    eneg
    0.25
    ä¸ĸ代
    0.25
    Activations Density 0.004%

    No Known Activations

    This feature has no known activations.