© 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. 43251
    Prev
    Next
    INDEX
    Explanations

    Code/Foreign Languages

    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
     judicial
    -0.28
    antry
    -0.26
     flip
    -0.26
    }'",
    -0.26
     Flip
    -0.26
     flatt
    -0.26
    OnUiThread
    -0.25
    ä»ī
    -0.25
     flipping
    -0.25
     flips
    -0.24
    POSITIVE LOGITS
    ieres
    0.28
    èĢIJ
    0.27
    ä¸įæĸŃ
    0.27
    ä¸įæĸŃåľ°
    0.27
    ä¸įåı¯
    0.26
    лан
    0.26
    è¾¾
    0.25
    iem
    0.25
    ATUS
    0.25
    Ja
    0.25
    Activations Density 0.000%

    No Known Activations