© 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. 63748
    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
     fon
    -0.27
    antal
    -0.27
     Temper
    -0.26
    templ
    -0.26
    least
    -0.25
    ickle
    -0.25
    appable
    -0.25
    tem
    -0.24
    æĥĬ
    -0.24
    æķ£
    -0.24
    POSITIVE LOGITS
    ControlItem
    0.25
     Ñģебе
    0.25
    acher
    0.25
    (Configuration
    0.24
    iciel
    0.24
    ssid
    0.23
    åĨħ马
    0.23
    acco
    0.23
    æ¼Ĥ
    0.23
     ourselves
    0.22
    Activations Density 62.319%

    No Known Activations

    This feature has no known activations.