© 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. 107895
    Prev
    Next
    INDEX
    Explanations

    case

    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
     Prev
    -0.25
     Farm
    -0.25
    è´¥
    -0.25
    ç»ħ
    -0.25
    IFE
    -0.25
    åģļ个
    -0.25
    eren
    -0.25
    ilos
    -0.24
     Wich
    -0.24
    çĿĢä»ĸ
    -0.24
    POSITIVE LOGITS
    isiert
    0.26
     corner
    0.25
    ãĤĤãģªãģı
    0.24
    è¿ĽéŨ
    0.24
    ita
    0.24
     summary
    0.24
    adas
    0.24
    ainer
    0.24
    æ¢ĥ
    0.24
    isi
    0.23
    Activations Density 2.396%

    No Known Activations