© 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. 19042
    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
     flat
    -0.28
    flat
    -0.26
    éľĩ
    -0.26
     sequencing
    -0.25
    atur
    -0.24
    é¢ł
    -0.24
    纵
    -0.24
    åIJĪ
    -0.24
    otation
    -0.24
     fat
    -0.24
    POSITIVE LOGITS
    为空
    0.28
    éĺ²ç©º
    0.28
     troll
    0.25
    å¹Į
    0.24
     homes
    0.24
    ondo
    0.24
    afb
    0.24
    åIJĥäºı
    0.24
    yne
    0.24
    æĹłå£°
    0.23
    Activations Density 0.119%

    No Known Activations