© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 25-TRANSCODER-HP
    7. 108146
    Prev
    Next
    INDEX
    Explanations

    code and databases

    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
    åĩŃ
    -0.28
    æĨij
    -0.27
    SourceType
    -0.27
    urs
    -0.25
    omi
    -0.25
    xies
    -0.25
    .BLL
    -0.24
    èĢħçļĦ
    -0.24
     Bav
    -0.24
    èľ
    -0.23
    POSITIVE LOGITS
    æł¼
    0.27
    æĪijå¸ĮæľĽ
    0.26
    è¶ĭåIJij
    0.26
     chÃło
    0.25
    าะ
    0.25
    ully
    0.25
    æIJľéĽĨ
    0.24
    æĦŁå®ĺ
    0.24
    ICLE
    0.24
    ument
    0.24
    Activations Density 0.133%

    No Known Activations