© 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. 111299
    Prev
    Next
    INDEX
    Explanations

    Code comments

    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
    Not in Any Lists

    No Comments

    Negative Logits
    raph
    -0.26
    ió
    -0.26
    xEB
    -0.26
    selectAll
    -0.26
    EB
    -0.25
     assuming
    -0.25
     setLocation
    -0.24
    anio
    -0.24
    øj
    -0.24
    VERRIDE
    -0.23
    POSITIVE LOGITS
    ä¸įæĸŃæī©å¤§
    0.31
    qu
    0.28
    ä»Ģä¹Īåij¢
    0.28
    ä»ĸå·²ç»ı
    0.26
    夷
    0.26
     powder
    0.25
    åİļåİļçļĦ
    0.24
     enlargement
    0.24
    åĪ¤
    0.24
    judge
    0.23
    Activations Density 0.022%

    No Known Activations