© 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. 52628
    Prev
    Next
    INDEX
    Explanations

    code and file paths

    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
    å¿«
    -0.28
     hell
    -0.28
     fast
    -0.27
    èIJ¥
    -0.26
    éĩį
    -0.25
    åłĨ
    -0.25
    fast
    -0.25
    æŃ£å½ĵ
    -0.25
     bur
    -0.24
    ê·¼
    -0.24
    POSITIVE LOGITS
    æľīæĿĥ
    0.29
     Voice
    0.28
    ä¸İåıijå±ķ
    0.28
    Voice
    0.27
    ÑĢоÑĪ
    0.26
     Sour
    0.25
    éħ¸
    0.25
     dams
    0.24
    åĻªéŁ³
    0.24
    CRE
    0.24
    Activations Density 0.024%

    No Known Activations