© 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. 130497
    Prev
    Next
    INDEX
    Explanations

    Writing and explanations

    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
    byn
    -0.30
    æīĵåİĭ
    -0.27
    strt
    -0.27
    eus
    -0.25
     Rosenberg
    -0.25
    åĮ¯
    -0.25
    èĥŀ
    -0.25
    çĭ¬ç«ĭ
    -0.24
    æĹ·
    -0.24
     sustain
    -0.24
    POSITIVE LOGITS
    off
    0.29
    åħ¬å¸ĥçļĦ
    0.29
    æݨåĩºçļĦ
    0.28
    roe
    0.27
    ereum
    0.26
    çłĤæµĨ
    0.26
     ly
    0.25
    çļĦèĦļæŃ¥
    0.25
    æĥ³ä¸įåĪ°
    0.25
    æıIJåĩºçļĦ
    0.25
    Activations Density 10.146%

    No Known Activations