© 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. 97237
    Prev
    Next
    INDEX
    Explanations

    ion

    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
     win
    -0.26
    ous
    -0.25
     fel
    -0.25
    ary
    -0.24
    zerbai
    -0.23
     thép
    -0.23
    _singleton
    -0.23
    ä¸ŃåįİæĸĩåĮĸ
    -0.23
     Datum
    -0.23
    .mods
    -0.23
    POSITIVE LOGITS
    å¹´åĨħ
    0.28
    å¾Ģå¹´
    0.27
     twentieth
    0.27
    ût
    0.25
    çľĭæ¸ħ
    0.25
    æĥħåĨµä¸ĭ
    0.24
    天èĬ±æĿ¿
    0.24
    æ·ĭ
    0.24
    tet
    0.23
    ä¸ĬéĿ¢
    0.23
    Activations Density 0.101%

    No Known Activations