© 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. 18343
    Prev
    Next
    INDEX
    Explanations

    Code/technical language 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
    ley
    -0.28
    åī¯æľ¬
    -0.27
     rejo
    -0.27
    inges
    -0.27
    ROME
    -0.26
    yg
    -0.26
    le
    -0.25
    lea
    -0.25
    ả
    -0.25
    让人们
    -0.25
    POSITIVE LOGITS
    adr
    0.35
    anas
    0.29
    dad
    0.28
    -quarter
    0.26
    æĦļ
    0.26
    èij©
    0.26
    å¨ľ
    0.25
    half
    0.25
    ania
    0.25
    Ĭ¶
    0.25
    Activations Density 19.078%

    No Known Activations