© 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. 19808
    Prev
    Next
    INDEX
    Explanations

    around

    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
    cctor
    -0.28
    itches
    -0.27
    çĹ£
    -0.25
    ossal
    -0.24
     ingl
    -0.24
    gars
    -0.24
    usters
    -0.23
    æīĵçł´
    -0.23
    å¡«è¡¥
    -0.23
    æIJIJ
    -0.22
    POSITIVE LOGITS
     around
    1.78
    around
    1.63
     Around
    1.62
    Around
    1.55
    -around
    1.28
    åĽ´ç»ķ
    1.23
     autour
    1.21
     round
    1.18
    round
    1.17
     вокÑĢÑĥг
    1.17
    Activations Density 28.575%

    No Known Activations