© 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. 62963
    Prev
    Next
    INDEX
    Explanations

    stand

    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
    åIJ¬çĿĢ
    -0.30
    æºIJ头
    -0.29
    (touch
    -0.28
    пиÑĤ
    -0.25
    RU
    -0.25
    jective
    -0.25
    è¶Ĭ
    -0.24
     oxy
    -0.24
    Runnable
    -0.23
    ä¸Ń
    -0.23
    POSITIVE LOGITS
     agora
    0.26
     vs
    0.26
     distinct
    0.26
     versus
    0.25
    nov
    0.25
     entail
    0.25
    çī¹æľīçļĦ
    0.25
    PostBack
    0.24
    AGE
    0.24
     Fore
    0.24
    Activations Density 18.929%

    No Known Activations