© 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. 42987
    Prev
    Next
    INDEX
    Explanations

    image description

    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
    Converted
    -0.26
    æİĸ
    -0.25
    åī
    -0.25
    ût
    -0.24
    resh
    -0.24
    çļĦåºĶçĶ¨
    -0.24
     Prop
    -0.24
    gü
    -0.24
    hea
    -0.23
     Bios
    -0.23
    POSITIVE LOGITS
     shown
    0.52
     presented
    0.40
     pictured
    0.40
     illustrated
    0.37
     depicted
    0.35
    shown
    0.35
     mentioned
    0.34
     discussed
    0.33
     outlined
    0.32
    æīĢ示
    0.32
    Activations Density 0.002%

    No Known Activations