© 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. 130927
    Prev
    Next
    INDEX
    Explanations

    code and references

    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
    Not in Any Lists

    No Comments

    Negative Logits
    cus
    -0.27
    幺
    -0.26
    å©Ĩ
    -0.26
    ä½İ声
    -0.25
    kos
    -0.25
    éĴĪ对æĢ§
    -0.25
    æĪĺçķ¥
    -0.25
    sonian
    -0.24
     tracer
    -0.23
    ÑĥÑĪ
    -0.23
    POSITIVE LOGITS
    lap
    0.26
    fortune
    0.25
    ajes
    0.25
    ãģ®ãģ¿
    0.25
     Beau
    0.24
    JO
    0.24
    讣
    0.24
    æĬĬä»ĸ
    0.24
     hayatı
    0.23
    uling
    0.23
    Activations Density 0.005%

    No Known Activations