© 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. 42827
    Prev
    Next
    INDEX
    Explanations

    comma/that

    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
    spin
    -0.30
     most
    -0.30
    most
    -0.29
     spin
    -0.28
     cheapest
    -0.28
    ëħĦ
    -0.27
     year
    -0.27
     spun
    -0.27
    (forms
    -0.27
     century
    -0.26
    POSITIVE LOGITS
     since
    0.29
    ä¸Ĭä¸ĭ
    0.28
    çĶ²
    0.28
     over
    0.27
    frican
    0.27
    ç®Ń
    0.27
     close
    0.26
    Another
    0.25
    rand
    0.24
    since
    0.24
    Activations Density 0.002%

    No Known Activations