© 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. 12-TRANSCODER-HP
    7. 22139
    Prev
    Next
    INDEX
    Explanations

    Names

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æľĢä½³
    -0.27
    æīĭ表
    -0.25
    ^(
    -0.25
    踮
    -0.25
    caller
    -0.24
     caller
    -0.24
    oser
    -0.23
    avig
    -0.23
    æ´¾åĩº
    -0.23
     (!(
    -0.23
    POSITIVE LOGITS
    ney
    0.28
    antal
    0.28
    utos
    0.27
     reinc
    0.27
    éĵļ
    0.27
    THREAD
    0.26
    å½¢æĪIJ
    0.25
    ë³ij
    0.25
    为ä»Ģä¹Īä¼ļ
    0.24
    å°¾
    0.24
    Activations Density 0.001%

    No Known Activations