© 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. 63700
    Prev
    Next
    INDEX
    Explanations

    abstract concepts

    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
    æ³¥
    -0.36
     Webb
    -0.27
    Qed
    -0.26
    åĹŁ
    -0.25
    åħ¬
    -0.25
    æĸľ
    -0.25
    çŃī级
    -0.24
    污泥
    -0.24
    ç«¿
    -0.24
    apel
    -0.23
    POSITIVE LOGITS
     bew
    0.26
     expresses
    0.26
    reation
    0.24
    /files
    0.23
     NATO
    0.23
    vio
    0.23
    cio
    0.23
    çĭį
    0.23
     livest
    0.23
     Mix
    0.23
    Activations Density 19.195%

    No Known Activations