© 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. 110916
    Prev
    Next
    INDEX
    Explanations

    )

    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
    ä¸įæĺ¯ä¸Ģ个
    -0.27
    ä¹ĭæĹħ
    -0.26
     (!((
    -0.26
    ä¸Ŀ毫
    -0.25
    ãĤĭãģĵãģ¨ãģ¯
    -0.25
    è¿Ļ份
    -0.25
     essentially
    -0.25
    åĬĽæ±Ĥ
    -0.24
    cul
    -0.24
    è¿Ľç¨ĭä¸Ń
    -0.24
    POSITIVE LOGITS
    reten
    0.28
    Ĥ¬
    0.27
     etc
    0.26
    ey
    0.26
    lan
    0.25
    ä¸ļ
    0.25
    etc
    0.25
    åģıè¿ľ
    0.24
     paran
    0.24
     lan
    0.23
    Activations Density 0.003%

    No Known Activations