© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 23-TRANSCODER-HP
    7. 162917
    Prev
    Next
    INDEX
    Explanations

    talking/discussing

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    好人
    -0.27
    éħįå¥Ĺ
    -0.27
     concent
    -0.27
    åĢĴ
    -0.26
    æµģéĢļ
    -0.26
    æĮģç»Ń
    -0.26
    æĪIJæľ¬
    -0.25
    æĸ¹æ³ķ
    -0.24
    æµģ
    -0.24
    æĢ»éĩı
    -0.24
    POSITIVE LOGITS
    就读
    0.29
    è°Īèµ·
    0.28
     secrecy
    0.27
    _fid
    0.26
    ãĤ·ãĥ¼
    0.26
    iese
    0.25
    бÑĢа
    0.25
    æīº
    0.25
    对å¾ħ
    0.25
    ä¿Ŀå¯Ĩ
    0.24
    Activations Density 0.001%

    No Known Activations