© 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. 142094
    Prev
    Next
    INDEX
    Explanations

    website excerpts

    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
    ındaki
    -0.30
     trần
    -0.27
    classNames
    -0.27
    uries
    -0.26
    ucle
    -0.26
    urable
    -0.26
    uous
    -0.25
    ince
    -0.25
    .ToInt
    -0.25
    ivant
    -0.25
    POSITIVE LOGITS
    nung
    0.31
     expect
    0.28
    Expect
    0.26
    (expect
    0.25
     Zucker
    0.25
    erra
    0.24
    è±¹
    0.24
    еÑģÑĤво
    0.24
    éľĩ
    0.24
    .consume
    0.24
    Activations Density 3.992%

    No Known Activations