© 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. 107732
    Prev
    Next
    INDEX
    Explanations

    general language

    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
    主è¦ģçĶ¨äºİ
    -0.28
    -guid
    -0.27
    cola
    -0.27
    è¿Ļåıªæĺ¯
    -0.26
    è¿Ļæĺ¯ä¸Ģç§į
    -0.26
    gles
    -0.26
    wargs
    -0.26
    æĺ¯ä¸ºäºĨ
    -0.25
    è¿ĻéĩĮæĺ¯
    -0.25
    yat
    -0.25
    POSITIVE LOGITS
    æ¢Ĺ
    0.27
    rends
    0.26
    enders
    0.26
    adio
    0.26
     Axe
    0.26
    Ĭ
    0.24
    æħ¨
    0.24
     anime
    0.24
    orie
    0.24
    _OFFSET
    0.24
    Activations Density 13.872%

    No Known Activations