© 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. 63693
    Prev
    Next
    INDEX
    Explanations

    they

    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
    çıı
    -0.26
    æĪijè¿ĺæĺ¯
    -0.26
    ÑĢÑĥÑĪ
    -0.26
    holder
    -0.26
    stay
    -0.25
    cop
    -0.25
    ograf
    -0.25
    “So
    -0.24
    tractor
    -0.24
     Go
    -0.24
    POSITIVE LOGITS
    _streams
    0.27
    adamente
    0.26
    keit
    0.25
    ilent
    0.25
    pot
    0.24
    严å¯Ĩ
    0.24
    Υ
    0.24
    que
    0.23
    å½ĵä¸Ń
    0.23
     pot
    0.23
    Activations Density 3.846%

    No Known Activations