© 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. 107900
    Prev
    Next
    INDEX
    Explanations

    stand

    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
    phem
    -0.26
    åIJĦ个çݯèĬĤ
    -0.25
    æĺ¯ä»İ
    -0.25
    åħ¶ä¸Ń
    -0.24
    å¤ļç§į
    -0.24
    èķĬ
    -0.23
    åħ¶ä¸Ńä¸Ģ个
    -0.23
    >Lorem
    -0.23
     slog
    -0.23
    fal
    -0.23
    POSITIVE LOGITS
     hydrated
    0.26
     straight
    0.24
    /current
    0.24
    åĪ·
    0.23
    好象
    0.23
    íĮIJ
    0.23
     matter
    0.23
    ведениÑı
    0.23
     up
    0.22
    oses
    0.22
    Activations Density 14.073%

    No Known Activations