© 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. 109429
    Prev
    Next
    INDEX
    Explanations

    encapsulation

    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
    社ä¼ļæķĪçĽĬ
    -0.30
    ame
    -0.27
    ised
    -0.26
    風
    -0.25
     wind
    -0.25
     Justice
    -0.24
    æĻĮ
    -0.24
    æ¦Ĥ念
    -0.24
    缮åħī
    -0.24
    roup
    -0.24
    POSITIVE LOGITS
     interchange
    0.28
    aries
    0.27
    nings
    0.27
     interchangeable
    0.25
    IntArray
    0.25
    naires
    0.24
    åĦ¿
    0.24
    éĴ¤
    0.24
    åħ«åį¦
    0.24
    æłĿ
    0.24
    Activations Density 1.198%

    No Known Activations