© 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. 20-TRANSCODER-HP
    7. 106961
    Prev
    Next
    INDEX
    Explanations

    Mathematical/scientific notation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Compression
    -0.28
    :Boolean
    -0.27
    æīĵæĸŃ
    -0.25
    èº²åľ¨
    -0.25
    compression
    -0.24
    ceeded
    -0.24
    compressed
    -0.24
    æĬ¢åħĪ
    -0.24
    -SA
    -0.23
    ispens
    -0.23
    POSITIVE LOGITS
    atrice
    0.25
    eft
    0.24
    åIJij举
    0.24
     zun
    0.24
    zeug
    0.24
    _XDECREF
    0.24
    choices
    0.24
    ares
    0.23
     Tune
    0.23
    quat
    0.23
    Activations Density 15.192%

    No Known Activations