© 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. 63114
    Prev
    Next
    INDEX
    Explanations

    Native

    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
     yap
    -0.26
    ou
    -0.26
    елÑı
    -0.25
    educt
    -0.25
     fl
    -0.25
     ou
    -0.24
     means
    -0.24
    abler
    -0.24
    Hur
    -0.24
    åĪļåĪļ
    -0.24
    POSITIVE LOGITS
    illac
    0.32
    ä¸įåłª
    0.30
    èĺħ
    0.26
    UGHT
    0.25
    BlockSize
    0.25
    azar
    0.25
    villa
    0.24
    ...]↵↵
    0.24
    gent
    0.24
     salesman
    0.24
    Activations Density 5.570%

    No Known Activations