© 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. 31-TRANSCODER-HP
    7. 18357
    Prev
    Next
    INDEX
    Explanations

    conversational/opinionated text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unexpected
    -0.29
    _cfg
    -0.26
    åı¯ä»¥ä»İ
    -0.26
    èĦ±
    -0.26
    ile
    -0.25
    æľĢ大çļĦ
    -0.25
    客æĪ·çļĦ
    -0.25
     Mig
    -0.24
    æĽ´å¤§çļĦ
    -0.24
    èĦ«
    -0.24
    POSITIVE LOGITS
     within
    0.39
    within
    0.34
    Within
    0.34
    elage
    0.31
    _within
    0.29
    æ´¼
    0.27
    åIJĪçIJĨ
    0.26
     Within
    0.26
    chers
    0.26
     reasonably
    0.26
    Activations Density 0.004%

    No Known Activations