© 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. 19-TRANSCODER-HP
    7. 8040
    Prev
    Next
    INDEX
    Explanations

    disclaimers and conflicts of interest

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å¼ĢçĿĢ
    -0.30
    è½»
    -0.29
    éĤ£æł·çļĦ
    -0.27
    è¿Ļæł·æīįèĥ½
    -0.26
    åĿĩè¡¡
    -0.26
     strs
    -0.26
    éĤ£æĺ¯
    -0.25
    èĵ¬
    -0.24
    è¿Ļç§į
    -0.24
     dicts
    -0.24
    POSITIVE LOGITS
    ivity
    0.26
    .mdl
    0.25
    iveness
    0.25
    ython
    0.25
    丢
    0.25
    dy
    0.25
    comes
    0.24
    inky
    0.24
     selections
    0.24
    kits
    0.24
    Activations Density 0.003%

    No Known Activations