© 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. 107087
    Prev
    Next
    INDEX
    Explanations

    Scientific/medical text

    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
    hare
    -0.26
    éĻĦ
    -0.25
     qa
    -0.24
    éĢĦ
    -0.24
    atitude
    -0.24
     Turtle
    -0.24
    {'
    -0.24
     carbon
    -0.23
    ä»ĺ
    -0.23
    çŀ§
    -0.23
    POSITIVE LOGITS
    -compatible
    0.27
     apparently
    0.25
    okers
    0.25
    -esque
    0.25
    帽
    0.24
    -style
    0.24
    -speaking
    0.24
    éŃĶæľ¯
    0.24
    ãģ©ãģĵãĤį
    0.24
    éĤ£åĦ¿
    0.24
    Activations Density 6.204%

    No Known Activations