© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 17754
    Prev
    Next
    INDEX
    Explanations

    Random text snippets

    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
    Negative Logits
    xa
    -0.30
    vore
    -0.27
    oki
    -0.26
    @a
    -0.26
    orer
    -0.25
    oder
    -0.25
    UGHT
    -0.24
    (named
    -0.24
    宸
    -0.24
    ocker
    -0.24
    POSITIVE LOGITS
     Cups
    0.27
    çķĮ
    0.26
    uel
    0.26
    egration
    0.25
    -setting
    0.25
    æł¼
    0.25
    cup
    0.25
     Bound
    0.25
    egr
    0.24
    æĸij
    0.24
    Activations Density 0.007%

    No Known Activations