© 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. 141880
    Prev
    Next
    INDEX
    Explanations

    software development

    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
     acceptable
    -0.32
     successfully
    -0.31
    è¶ĬæĿ¥è¶Ĭ好
    -0.30
     perfect
    -0.30
    满æĦı
    -0.29
    åIJĪæł¼
    -0.29
     okay
    -0.28
    esome
    -0.28
    isOk
    -0.28
    oka
    -0.28
    POSITIVE LOGITS
    èĩªåıij
    0.32
    æĢ»æľī
    0.26
    诱
    0.25
    ensible
    0.25
    ADATA
    0.25
    cycl
    0.24
    çĶ¨æĿ¥
    0.24
    åıijçİ°
    0.24
    chez
    0.23
    èĩªåѦ
    0.23
    Activations Density 64.298%

    No Known Activations