© 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. 107003
    Prev
    Next
    INDEX
    Explanations

    and

    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
     Nou
    -0.28
     important
    -0.27
     importance
    -0.26
    éĿŀ常éĩįè¦ģ
    -0.26
    itle
    -0.26
    éĩįè¦ģ
    -0.26
     Soon
    -0.26
    ä¸ī人
    -0.25
     essential
    -0.25
    è¿Ļè¯Ŀ
    -0.25
    POSITIVE LOGITS
     Scalars
    0.28
     vets
    0.27
    .SDK
    0.26
    AVIS
    0.26
    .Sin
    0.25
    @js
    0.25
    ROME
    0.25
    $class
    0.25
    xCB
    0.24
    ä¿Ŀå®ī
    0.24
    Activations Density 0.525%

    No Known Activations