© 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. 112947
    Prev
    Next
    INDEX
    Explanations

    a

    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
     dele
    -0.29
    tok
    -0.25
     ties
    -0.24
    ift
    -0.24
    ificant
    -0.23
    hil
    -0.23
    ivist
    -0.23
    ques
    -0.23
     furn
    -0.23
    容纳
    -0.23
    POSITIVE LOGITS
    第ä¸ĢæĿ¡
    0.30
    è¿ŀ线
    0.27
    åı£å¤´
    0.26
    輪
    0.25
    née
    0.25
     "")↵↵
    0.25
    产åĵģçłĶåıij
    0.24
    该项
    0.24
    aroo
    0.24
     !_
    0.23
    Activations Density 0.021%

    No Known Activations