© 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. 19-TRANSCODER-HP
    7. 9620
    Prev
    Next
    INDEX
    Explanations

    comes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    eses
    -0.24
    ãģ¹ãģį
    -0.23
    edList
    -0.23
    _literals
    -0.23
     bound
    -0.23
    ritos
    -0.23
     form
    -0.23
    enk
    -0.23
    Ïį
    -0.23
    gression
    -0.22
    POSITIVE LOGITS
    è¦ģä¹Ī
    0.29
    GIN
    0.27
    被æīĵ
    0.26
     schl
    0.25
    flater
    0.25
    åı¯æĥ³èĢĮçŁ¥
    0.24
     kull
    0.24
    peek
    0.24
     insanely
    0.24
     Gio
    0.23
    Activations Density 0.434%

    No Known Activations