© 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. 32-TRANSCODER-HP
    7. 43898
    Prev
    Next
    INDEX
    Explanations

    say "?"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ever
    -0.32
     ahead
    -0.32
    aida
    -0.31
    vine
    -0.30
     next
    -0.30
    æĢ§çļĦ
    -0.30
    æĢ§åĴĮ
    -0.29
     future
    -0.29
     notably
    -0.29
     this
    -0.28
    POSITIVE LOGITS
    ç½ijç»ľä¼łæĴŃ
    0.33
    mania
    0.29
     Crushers
    0.26
    çıį
    0.25
    Approved
    0.25
    alus
    0.24
    åĬ¿å¤´
    0.24
    <?↵
    0.24
    би
    0.24
     !↵
    0.24
    Activations Density 0.005%

    No Known Activations