© 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. 62560
    Prev
    Next
    INDEX
    Explanations

    6

    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
    |-
    -0.29
     luk
    -0.27
    }*
    -0.27
     Unsure
    -0.26
    èĥĮæĻ¯ä¸ĭ
    -0.26
    .experimental
    -0.26
     ATV
    -0.25
    AAP
    -0.25
    gren
    -0.25
    åīĬå¼±
    -0.24
    POSITIVE LOGITS
     proportion
    0.30
     yapıl
    0.27
    秣
    0.25
    æľīæĦı
    0.25
     percentage
    0.25
     parties
    0.25
    å½ĵäºĭ
    0.25
    æ¯Ķä¾ĭ
    0.24
    áĢļ
    0.24
     ник
    0.24
    Activations Density 15.188%

    No Known Activations