© 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. 32-TRANSCODER-HP
    7. 42237
    Prev
    Next
    INDEX
    Explanations

    code copyright notices

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     book
    -0.41
    _book
    -0.39
    书
    -0.37
     Book
    -0.37
    书ä¸Ń
    -0.36
    -book
    -0.36
    .book
    -0.35
    book
    -0.35
    Book
    -0.33
     Reserve
    -0.33
    POSITIVE LOGITS
     settled
    0.30
     Cla
    0.30
     aborted
    0.27
     resolved
    0.27
     rejected
    0.26
     Jal
    0.26
    -scripts
    0.25
     answered
    0.24
     ens
    0.24
    ãģĭãĤĬ
    0.24
    Activations Density 0.015%

    No Known Activations