© 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. 42900
    Prev
    Next
    INDEX
    Explanations

    tell

    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
    ä¸Ģ个
    -0.27
    çļĦéĤ£ä¸ª
    -0.26
    HC
    -0.26
    ä»İ
    -0.24
    posted
    -0.24
    wat
    -0.24
     ...
    -0.24
    hatt
    -0.24
    æĬĬè¿Ļ个
    -0.24
    ene
    -0.23
    POSITIVE LOGITS
     employers
    0.36
     parents
    0.35
     people
    0.35
     elders
    0.34
    人们
    0.34
     adults
    0.33
     grandparents
    0.32
    èĢģå¸Ī们
    0.31
     therapists
    0.31
     users
    0.31
    Activations Density 0.014%

    No Known Activations