© 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. 19373
    Prev
    Next
    INDEX
    Explanations

    early and ear

    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
    angep
    -0.29
    isted
    -0.27
    apot
    -0.25
     Fortress
    -0.25
    éd
    -0.25
    çŃĶ辩
    -0.25
    /by
    -0.25
    åľ¨åŃ¦æł¡
    -0.25
     fortress
    -0.24
    aten
    -0.23
    POSITIVE LOGITS
    ðŁĽł
    0.27
    mill
    0.26
    æľŁ
    0.26
    SPATH
    0.26
    lette
    0.25
    åIJ§
    0.25
    å°±å¤ŁäºĨ
    0.25
    æīĴ
    0.25
    马ä¸ģ
    0.25
    满äºĨ
    0.25
    Activations Density 0.072%

    No Known Activations