© 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. 129593
    Prev
    Next
    INDEX
    Explanations

    iph

    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
    stri
    -0.29
    aug
    -0.28
    ali
    -0.26
    åĵŃäºĨ
    -0.25
    æŃ£åĢ¼
    -0.25
    격
    -0.25
    mo
    -0.25
     moi
    -0.24
    laden
    -0.24
    交æ±ĩ
    -0.24
    POSITIVE LOGITS
    为广大
    0.29
    çĦĺ
    0.28
     Kore
    0.28
    容
    0.27
    -tip
    0.26
    ä¹Łæĺ¯ä¸Ģ个
    0.25
    æĸ°çļĦä¸Ģå¹´
    0.25
    æĸ°å¼Ģ
    0.25
     Bell
    0.25
    éģĵ
    0.25
    Activations Density 0.593%

    No Known Activations