© 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. 107272
    Prev
    Next
    INDEX
    Explanations

    sometimes

    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
    Not in Any Lists

    No Comments

    Negative Logits
    interop
    -0.32
    okers
    -0.26
    éĩįåŀĭ
    -0.26
    =(↵
    -0.26
     поÑĤеÑĢÑı
    -0.25
    æľ«
    -0.24
    agra
    -0.24
    itters
    -0.24
    -chart
    -0.24
    åı¯ä»¥çĽ´æİ¥
    -0.24
    POSITIVE LOGITS
     also
    0.32
     English
    0.32
     translated
    0.30
     singular
    0.29
     abbreviated
    0.29
    èĭ±è¯Ń
    0.29
     called
    0.28
    çŁ£
    0.28
     translates
    0.28
     english
    0.28
    Activations Density 0.250%

    No Known Activations