© 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. 128905
    Prev
    Next
    INDEX
    Explanations

    conflict

    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
    -0.26
    TextChanged
    -0.26
    ä¸ĢæĹı
    -0.24
    autos
    -0.23
    ?page
    -0.23
    åįħ
    -0.22
     filetype
    -0.22
     Perm
    -0.22
    atha
    -0.22
    ä»Ĺ
    -0.22
    POSITIVE LOGITS
    ões
    0.34
    ous
    0.31
    wise
    0.29
    eting
    0.29
    ive
    0.29
    us
    0.28
    less
    0.26
    éĩįéĩį
    0.26
     succeeding
    0.26
    se
    0.25
    Activations Density 0.093%

    No Known Activations