© 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. 31-TRANSCODER-HP
    7. 18715
    Prev
    Next
    INDEX
    Explanations

    Medicine and medical research

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è´¨
    -0.30
    ackers
    -0.27
    çŁ¿
    -0.26
    æĿŁ
    -0.26
    /sdk
    -0.25
    éħ¿
    -0.25
     Trades
    -0.25
     bam
    -0.25
     Lesb
    -0.25
    æľ«
    -0.25
    POSITIVE LOGITS
    entials
    0.27
    éŃĶ
    0.25
    失败
    0.25
     kurz
    0.24
    nar
    0.24
    æİĴåIJį第
    0.24
     Mane
    0.24
    GAN
    0.24
    åį«è§Ĩ
    0.24
     Gov
    0.24
    Activations Density 0.014%

    No Known Activations