© 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. 33-TRANSCODER-HP
    7. 54272
    Prev
    Next
    INDEX
    Explanations

    Animals and environment

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    zell
    -0.29
    ari
    -0.28
    声
    -0.27
    å±ŀå®ŀ
    -0.26
    åIJ¦
    -0.26
     decals
    -0.25
     treatment
    -0.25
    å¾®
    -0.25
    z
    -0.25
    copies
    -0.23
    POSITIVE LOGITS
     sand
    0.30
    èĬĴ
    0.29
     North
    0.28
    赤
    0.28
    North
    0.28
    åĪ©çī©
    0.28
     g
    0.27
     red
    0.27
     will
    0.27
     unp
    0.27
    Activations Density 0.020%

    No Known Activations