© 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. 64031
    Prev
    Next
    INDEX
    Explanations

    Code snippets, text segments

    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
    YGON
    -0.27
    åĪĨãģĭãĤĭ
    -0.26
    触
    -0.23
    临港
    -0.23
     spoilers
    -0.22
    æ¼Ķç»İ
    -0.22
    sburgh
    -0.22
    defgroup
    -0.22
    ême
    -0.22
    ãĤıãģĭãĤĭ
    -0.21
    POSITIVE LOGITS
     fart
    0.25
     fclose
    0.25
    è´®
    0.23
    éĻĽ
    0.23
     pooling
    0.23
     Tem
    0.23
    pile
    0.23
    blr
    0.23
    æĥ
    0.23
     comput
    0.23
    Activations Density 86.358%

    No Known Activations