© 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. 22-TRANSCODER-HP
    7. 129064
    Prev
    Next
    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Dod
    -0.28
     obsess
    -0.26
    velle
    -0.26
    æľīèĥ½åĬĽ
    -0.25
    æľīåı¯èĥ½
    -0.25
    rief
    -0.25
    odie
    -0.24
    çĽ¡
    -0.24
     loophole
    -0.24
    upid
    -0.24
    POSITIVE LOGITS
    acket
    0.28
    线ä¸ĭ
    0.25
     y
    0.25
    éĶ®
    0.25
    äºĨ
    0.24
     letters
    0.24
    åŃĹ
    0.23
    E
    0.23
    eresa
    0.23
    æľŁ
    0.23
    Activations Density 0.008%

    No Known Activations