© 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. 63050
    Prev
    Next
    INDEX
    Explanations

    disease

    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
     —
    -0.28
    åĭĥ
    -0.27
    !↵
    -0.26
    bold
    -0.26
    yen
    -0.26
    _fb
    -0.25
    variation
    -0.25
     fö
    -0.25
     cheers
    -0.24
    vec
    -0.24
    POSITIVE LOGITS
    ought
    0.29
    ]";↵
    0.25
     Gdk
    0.24
    STALL
    0.24
    åIJĦæł·
    0.24
    ikan
    0.24
    æł¼å¤ĸ
    0.24
    terms
    0.23
    æĥ§
    0.23
    KeyId
    0.23
    Activations Density 78.356%

    No Known Activations