© 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. 130687
    Prev
    Next
    INDEX
    Explanations

    academic/technical texts

    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
    å¿ħè¦ģ
    -0.33
     medi
    -0.25
    ç¡®
    -0.25
    Filters
    -0.25
    idor
    -0.24
    çģ¿
    -0.24
    emer
    -0.24
    æļ®
    -0.24
    undi
    -0.23
    otomy
    -0.23
    POSITIVE LOGITS
     twist
    0.28
    agra
    0.28
    ķĮ
    0.26
     purposes
    0.25
    代表çĿĢ
    0.24
    .Circle
    0.24
    çķĮ
    0.24
    érica
    0.24
     scaled
    0.24
    ãģĵãģ¨ãģ«
    0.24
    Activations Density 68.620%

    No Known Activations