© 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. 44005
    Prev
    Next
    INDEX
    Explanations

    processing

    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
    Negative Logits
     processing
    -0.71
     Processing
    -0.68
    Processing
    -0.65
    å¤ĦçIJĨ
    -0.63
    _processing
    -0.59
     processed
    -0.59
     Processor
    -0.56
    èĻķçIJĨ
    -0.56
    åĬłå·¥
    -0.56
    processing
    -0.55
    POSITIVE LOGITS
    毽
    0.28
    ä¹ĭä½Ļ
    0.27
    units
    0.26
    åįķåħĥ
    0.26
    unit
    0.25
    schools
    0.25
     Freeman
    0.24
     зло
    0.24
     Jad
    0.24
     UIF
    0.24
    Activations Density 0.017%

    No Known Activations