© 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. 30-TRANSCODER-HP
    7. 96892
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    APO
    -0.27
    ä¸ī个
    -0.27
    ä¸ī次
    -0.27
    两次
    -0.27
    æī¹åĩĨ
    -0.26
    åįł
    -0.25
    åĽĽä¸ª
    -0.25
    åı¬
    -0.25
    arLayout
    -0.25
     cues
    -0.24
    POSITIVE LOGITS
     massive
    0.32
     nave
    0.30
    huge
    0.29
     colossal
    0.29
     huge
    0.29
     sides
    0.28
     enormous
    0.28
     entrances
    0.27
     retaining
    0.27
     giant
    0.26
    Activations Density 0.889%

    No Known Activations