© 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. 30-TRANSCODER-HP
    7. 95033
    Prev
    Next
    INDEX
    Explanations

    ynes

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    åħļåĴĮåĽ½å®¶
    -0.28
    æĮijåīĶ
    -0.25
     rtrim
    -0.24
    ä¸ŃåĽ½ç»ıæµİ
    -0.24
    chin
    -0.24
    å§Ķåijĺ
    -0.24
    å¹²èĦĨ
    -0.23
     SEN
    -0.23
     Kart
    -0.23
    ello
    -0.23
    POSITIVE LOGITS
    åıį
    0.30
     verbose
    0.27
    æĬ«éľ²
    0.27
    æŃ£
    0.27
    eties
    0.27
    æŀ¢
    0.26
    leness
    0.26
    åłĨ
    0.26
    羸
    0.25
    éĤ¸
    0.25
    Activations Density 0.002%

    No Known Activations