© 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. 18529
    Prev
    Next
    INDEX
    Explanations

    code/names/phrases

    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
     biên
    -0.31
    _DF
    -0.26
    elin
    -0.25
    пл
    -0.25
    æĻºåĬĽ
    -0.25
    å¹´ä¹ĭ
    -0.25
     DF
    -0.24
    exc
    -0.23
    æ±Ģ
    -0.23
    æĽ²çº¿
    -0.23
    POSITIVE LOGITS
     Giles
    0.28
    ionale
    0.26
    aris
    0.26
    umont
    0.26
    ahoma
    0.26
    haft
    0.26
    ania
    0.25
    ownik
    0.25
     added
    0.25
    gars
    0.25
    Activations Density 0.050%

    No Known Activations