© 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. 107865
    Prev
    Next
    INDEX
    Explanations

    business and legal documents

    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
    çİ°å®ŀ
    -0.27
    æ¶Įçİ°
    -0.26
    anj
    -0.25
    mah
    -0.25
     encounters
    -0.25
    wick
    -0.25
     props
    -0.24
    çı¾å¯¦
    -0.24
    appiness
    -0.24
     enjoyment
    -0.23
    POSITIVE LOGITS
    clo
    0.30
     clo
    0.28
    atura
    0.28
     pairwise
    0.25
    à¸ģาย
    0.24
     Harding
    0.24
    <!--[
    0.23
     McCart
    0.23
    .createFrom
    0.23
    åı£æ°Ķ
    0.23
    Activations Density 1.947%

    No Known Activations