© 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. 42353
    Prev
    Next
    INDEX
    Explanations

    Object recognition

    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
    kJ
    -0.27
    æ¯Ĵç´ł
    -0.26
    antes
    -0.26
    éŃĶçİĭ
    -0.26
     belts
    -0.26
     vectors
    -0.25
    kre
    -0.25
    iga
    -0.25
     angels
    -0.25
     Angels
    -0.24
    POSITIVE LOGITS
     object
    0.77
    object
    0.70
     Object
    0.66
    -object
    0.61
    _object
    0.61
    对象
    0.59
    Object
    0.58
     computer
    0.58
    	object
    0.57
    ForObject
    0.57
    Activations Density 0.016%

    No Known Activations