© 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. 129884
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    éģĵ
    -0.32
    æģ©
    -0.26
    parsers
    -0.25
    ICS
    -0.24
    gh
    -0.24
    oca
    -0.24
    gies
    -0.23
    agh
    -0.23
     Oc
    -0.23
    -k
    -0.23
    POSITIVE LOGITS
    pered
    0.29
    versed
    0.26
    æĸľ
    0.26
    ingles
    0.25
    å¿ĥå¢ĥ
    0.25
    lique
    0.25
    ä¸Ģç»ı
    0.25
    åºĬä½į
    0.25
    ople
    0.24
     conscious
    0.24
    Activations Density 0.119%

    No Known Activations

    This feature has no known activations.