© 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. 42046
    Prev
    Next
    INDEX
    Explanations

    1. Introduction

    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
    otre
    -0.29
    0
    -0.27
    åıĮ
    -0.27
    asse
    -0.27
    é¦Ļ
    -0.25
    代
    -0.25
    omon
    -0.24
    	catch
    -0.24
    éĻ¤å¤ĸ
    -0.23
    ÙħÙĬÙĨ
    -0.23
    POSITIVE LOGITS
    iate
    0.26
    转载请
    0.25
    ÙģÙĪ
    0.25
    APPED
    0.25
    èģĮä¸ļæĬĢæľ¯
    0.24
    izio
    0.24
    UTE
    0.24
    รา
    0.24
    resident
    0.24
    uat
    0.24
    Activations Density 0.001%

    No Known Activations