© 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. 107787
    Prev
    Next
    INDEX
    Explanations

    technical/scientific 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
    çŁ¥æĥħ
    -0.28
     misinformation
    -0.27
    tees
    -0.27
    stown
    -0.25
    EDIA
    -0.25
     sign
    -0.25
     Bengal
    -0.25
    çŃ¾
    -0.25
     ****************************************************************
    -0.25
    ä¸įæŃ»
    -0.24
    POSITIVE LOGITS
    è¿ĩ渡
    0.28
    åģļçĶŁæĦı
    0.28
    çĶ¨äºİ
    0.26
     cit
    0.25
    -bl
    0.25
    cit
    0.25
    å¢ħ
    0.25
    à¤Ī
    0.24
     vive
    0.24
    ä¸ĬåѦ
    0.24
    Activations Density 61.641%

    No Known Activations