© 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. 107195
    Prev
    Next
    INDEX
    Explanations

    football

    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
    åıĤ
    -0.25
    åħ¼
    -0.24
    OfDay
    -0.24
    -neck
    -0.23
    лÑĥÑĩ
    -0.23
    åħ¬åħ±
    -0.23
    åĪĨæĶ¯
    -0.23
    éħĬ
    -0.22
     {}),↵
    -0.22
    åħ¬åħ±æľįåĬ¡
    -0.22
    POSITIVE LOGITS
    er
    0.36
    HING
    0.34
    ing
    0.31
    ity
    0.30
    bett
    0.29
    matic
    0.29
    åīįèĢħ
    0.28
    sold
    0.27
    es
    0.26
    ÑģÑĤва
    0.26
    Activations Density 1.359%

    No Known Activations