© 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. 43583
    Prev
    Next
    INDEX
    Explanations

    data

    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
    erior
    -0.29
    atables
    -0.29
    icipants
    -0.28
    ihilation
    -0.28
    acters
    -0.28
    iration
    -0.28
    sent
    -0.27
    Łèĥ½
    -0.27
    itions
    -0.26
    entions
    -0.26
    POSITIVE LOGITS
     set
    0.38
    (set
    0.32
     setzen
    0.29
     martin
    0.28
    )set
    0.28
    =set
    0.28
    	set
    0.27
     setName
    0.27
    setData
    0.27
    aho
    0.26
    Activations Density 0.010%

    No Known Activations