© 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. 20-TRANSCODER-HP
    7. 107817
    Prev
    Next
    INDEX
    Explanations

    Factual/informative texts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     interception
    -0.26
     incidental
    -0.26
    periments
    -0.26
    ipher
    -0.25
     micron
    -0.24
    ç«ŀèµĽ
    -0.24
     intermedi
    -0.23
     mitt
    -0.23
     defenses
    -0.23
    广大群ä¼Ĺ
    -0.23
    POSITIVE LOGITS
    åİĨåı²
    0.66
     historical
    0.57
     history
    0.54
    çļĦåİĨåı²
    0.54
    æŃ·åı²
    0.52
     Historical
    0.49
    histor
    0.48
     historic
    0.47
     histor
    0.46
     histories
    0.46
    Activations Density 89.953%

    No Known Activations