© 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. 27-TRANSCODER-HP
    7. 56492
    Prev
    Next
    INDEX
    Explanations

    ij

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     diffuse
    -0.31
    ç½ij讯
    -0.28
    åĴ¨
    -0.27
    çľģéĴ±
    -0.26
    scatter
    -0.24
     Peg
    -0.24
     ÑĢаз
    -0.24
    mdir
    -0.24
     roam
    -0.24
     scatter
    -0.24
    POSITIVE LOGITS
    æ®Ĭ
    0.28
    /vendors
    0.27
    episode
    0.26
    è´¼
    0.26
    ild
    0.26
    ç½Ł
    0.26
    éĻĢ
    0.25
    ngrx
    0.25
    awns
    0.24
    羣çIJĨ
    0.24
    Activations Density 0.003%

    No Known Activations