© 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. 88620
    Prev
    Next
    INDEX
    Explanations

    copyright-related text

    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
     tempted
    -0.09
    rase
    -0.08
    rary
    -0.08
    bez
    -0.08
    åĿ¦
    -0.08
     thunk
    -0.07
    reuse
    -0.07
    maf
    -0.07
    tempt
    -0.07
    .cur
    -0.07
    POSITIVE LOGITS
     field
    0.09
     Sound
    0.08
    ο
    0.08
     MC
    0.08
     Nut
    0.08
    wards
    0.08
     sound
    0.08
    olland
    0.08
    ìłģìĿ´
    0.08
     Sen
    0.08
    Activations Density 0.001%

    No Known Activations