© 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. 31-TRANSCODER-HP
    7. 17964
    Prev
    Next
    INDEX
    Explanations

    say capitalized sentence starters

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     nạn
    -0.27
     wast
    -0.27
    —even
    -0.26
    .Compiler
    -0.26
     Territories
    -0.25
    _INLINE
    -0.24
    æĹłäºº
    -0.24
    åħľ
    -0.24
    (para
    -0.24
    åı¯ä»¥è¯´
    -0.24
    POSITIVE LOGITS
     Alright
    0.37
     Okay
    0.36
     That
    0.35
     Wait
    0.35
    Okay
    0.34
    Wait
    0.33
     See
    0.32
    Alright
    0.32
     trust
    0.32
    _that
    0.32
    Activations Density 0.906%

    No Known Activations