© 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. 25-TRANSCODER-HP
    7. 109396
    Prev
    Next
    INDEX
    Explanations

    all or of

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    Ark
    -0.26
    contents
    -0.24
    享
    -0.24
    ele
    -0.24
     Contents
    -0.23
    æľĭåıĭåľĪ
    -0.23
    sel
    -0.23
     contents
    -0.23
    è´ª
    -0.23
    Õª
    -0.23
    POSITIVE LOGITS
    è¿Ļä¸ĢçĤ¹
    0.50
    è¿ĻçĤ¹
    0.44
    è¿Ļ件äºĭæĥħ
    0.41
    è¿Ļäºĭ
    0.40
     this
    0.40
    è¿Ļ对äºİ
    0.39
    è¿Ļ件äºĭ
    0.38
    è¿Ļæł·åģļ
    0.38
    è¿Ļç§įæĥħåĨµ
    0.36
    è¿Ļ个
    0.35
    Activations Density 1.022%

    No Known Activations