© 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. 33-TRANSCODER-HP
    7. 52601
    Prev
    Next
    INDEX
    Explanations

    villains

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æł¡
    -0.28
     borderTop
    -0.28
    ialis
    -0.27
    orld
    -0.27
    ulant
    -0.26
    ateau
    -0.25
    ajaran
    -0.25
    åħīçݯ
    -0.25
    èĤļ
    -0.25
     Universe
    -0.25
    POSITIVE LOGITS
     spies
    0.29
    åĬ³
    0.29
     characters
    0.28
     invaders
    0.27
     forces
    0.27
     adversaries
    0.27
     attackers
    0.27
     forests
    0.26
     Ses
    0.26
    ecess
    0.26
    Activations Density 0.057%

    No Known Activations