© 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. 20-TRANSCODER-HP
    7. 129102
    Prev
    Next
    INDEX
    Explanations

    Dan

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Celt
    -0.26
    aber
    -0.25
    ncy
    -0.24
     ÑģоÑĤ
    -0.24
    é¦Ĩ
    -0.23
    asher
    -0.23
     memberId
    -0.23
     figure
    -0.23
    ØŃضر
    -0.22
     quot
    -0.22
    POSITIVE LOGITS
    orgen
    0.29
    æīĵçł´
    0.28
    íĶ½
    0.27
    лиÑģÑĤ
    0.25
    大大
    0.25
    eward
    0.25
    æī©å±ķ
    0.24
    åĬŁå¾·
    0.24
    /renderer
    0.24
    à¸Ķวà¸ĩ
    0.24
    Activations Density 0.282%

    No Known Activations