© 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. 63960
    Prev
    Next
    INDEX
    Explanations

    chat bot 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
     Kickstarter
    -0.27
     Matth
    -0.25
    yc
    -0.25
     guessed
    -0.25
    aser
    -0.24
    Verifier
    -0.24
     choÄĩ
    -0.23
     Belgi
    -0.23
    amera
    -0.23
    usta
    -0.23
    POSITIVE LOGITS
    iens
    0.29
    ff
    0.27
    ÙĪÙĥ
    0.26
     ff
    0.25
     TA
    0.25
    Ø·ÙĦ
    0.24
    çļĦè¿ĽæŃ¥
    0.24
    Ñģла
    0.24
    èļª
    0.24
    çļĦçĶŁåij½
    0.24
    Activations Density 9.654%

    No Known Activations