© 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. 141189
    Prev
    Next
    INDEX
    Explanations

    acquisition

    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
    ç¿ķ
    -0.27
     tô
    -0.26
    aison
    -0.26
    ourke
    -0.26
    isé
    -0.26
     Unter
    -0.26
    ä¸ī人
    -0.25
    ãĤĴåħ¥ãĤĮ
    -0.25
     pour
    -0.24
    ligt
    -0.24
    POSITIVE LOGITS
     Deliver
    0.28
    OrDefault
    0.27
    bü
    0.26
    umph
    0.26
    quired
    0.25
     carved
    0.25
    amar
    0.25
    impl
    0.25
    from
    0.24
    FO
    0.24
    Activations Density 12.707%

    No Known Activations