© 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. 110974
    Prev
    Next
    INDEX
    Explanations

    Past tense verbs

    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
     collapsed
    -0.26
    rang
    -0.26
    ween
    -0.25
    .cum
    -0.25
    Assembler
    -0.25
    ingham
    -0.24
    Shock
    -0.24
    声éģĵ
    -0.24
    contri
    -0.23
    åı¦ä¸Ģ个
    -0.23
    POSITIVE LOGITS
     Finch
    0.25
    vertime
    0.25
    æ¾Į
    0.25
    lit
    0.24
    INET
    0.24
    _globals
    0.24
    身份
    0.24
    åħļåı²åŃ¦ä¹ł
    0.24
    it
    0.24
    itzer
    0.24
    Activations Density 0.001%

    No Known Activations