© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 41637
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    ter
    -0.27
     transcription
    -0.27
     removal
    -0.27
    æĤ¬æĮĤ
    -0.26
     elimination
    -0.26
    ãģĭãģijãĤĭ
    -0.26
    ewood
    -0.25
    hort
    -0.25
     enrichment
    -0.25
    ound
    -0.25
    POSITIVE LOGITS
     grayscale
    0.31
    ACS
    0.28
    赶æĿ¥
    0.26
    çģ°èī²
    0.26
    DVD
    0.26
    çģ°
    0.26
     initialised
    0.25
    (dec
    0.25
    åľ¨çĶŁæ´»ä¸Ń
    0.25
    åİĭåĬĽ
    0.25
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.