© 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. 18-TRANSCODER-HP
    7. 88972
    Prev
    Next
    INDEX
    Explanations

    News articles, names

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uffs
    -0.28
    åį·
    -0.27
    åij¨å²ģ
    -0.26
    volt
    -0.26
    **↵
    -0.26
    ________
    -0.25
     rolls
    -0.25
    æīŃ转
    -0.25
     **
    -0.25
    **
    -0.25
    POSITIVE LOGITS
    .preference
    0.29
    æģ°æģ°
    0.26
    æĥħ人
    0.24
    éĦī
    0.24
    icast
    0.23
    ularity
    0.23
    åĢĴæĺ¯
    0.23
    ãģ¯ãģĤãĤĭ
    0.23
    乡éķĩ
    0.23
    )$/
    0.23
    Activations Density 29.059%

    No Known Activations