© 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. 19755
    Prev
    Next
    INDEX
    Explanations

    Stan/an

    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
    eties
    -0.29
    eba
    -0.27
    gom
    -0.26
    åīįæīĢ
    -0.26
    å¤©åľ°
    -0.26
    _DISK
    -0.25
    ATIO
    -0.24
    Entropy
    -0.24
    EMY
    -0.24
    IRST
    -0.24
    POSITIVE LOGITS
    los
    0.30
    dra
    0.29
    LOS
    0.29
    æĻĤéĸĵ
    0.28
    au
    0.28
    cion
    0.26
    æĸ¯é¡¿
    0.25
    ystone
    0.25
    Time
    0.25
     dÃ¼ÅŁ
    0.25
    Activations Density 0.016%

    No Known Activations