© 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. 63920
    Prev
    Next
    INDEX
    Explanations

    general text

    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
     Apart
    -0.30
     apart
    -0.29
     archetype
    -0.26
    åĢĴ
    -0.26
     i
    -0.26
    i
    -0.26
    CK
    -0.26
     wed
    -0.26
    Apart
    -0.26
     aside
    -0.25
    POSITIVE LOGITS
    htag
    0.28
    htags
    0.28
    æ¢ħèĬ±
    0.26
    çļĦæ¦Ĥçİĩ
    0.26
    $route
    0.26
    äºī夺
    0.26
     sanct
    0.26
    hips
    0.26
    --č↵
    0.26
    çľĭäºĨä¸Ģ
    0.25
    Activations Density 13.398%

    No Known Activations