© 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. 19199
    Prev
    Next
    INDEX
    Explanations

    Code snippets

    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
    ese
    -0.27
     independent
    -0.27
    ey
    -0.26
    Independent
    -0.25
    ä¸įåıĺ
    -0.24
    opsis
    -0.24
    -independent
    -0.24
    éļĶ
    -0.23
    vol
    -0.23
    QUE
    -0.23
    POSITIVE LOGITS
     Rare
    0.31
    æ»´æ»´
    0.28
    ooter
    0.27
    OMIC
    0.26
    ouv
    0.25
     repetitions
    0.25
     propor
    0.25
    éĩįçİ°
    0.24
    uru
    0.23
     Dou
    0.23
    Activations Density 0.890%

    No Known Activations