© 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. 63561
    Prev
    Next
    INDEX
    Explanations

    Locations

    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
    tml
    -0.28
    æĸ°è¥¿
    -0.27
    pak
    -0.27
     Maced
    -0.27
    桦
    -0.27
    æĢĽ
    -0.26
    éĹ²ç½®
    -0.26
     programma
    -0.25
    _ORD
    -0.25
    åĵĪ
    -0.24
    POSITIVE LOGITS
    çĶ¸
    0.34
    opper
    0.26
    apter
    0.25
    riority
    0.24
    ìłľ
    0.24
    渡
    0.23
     Criteria
    0.23
    Contin
    0.23
    æĤ¦
    0.23
     æŁ¥
    0.22
    Activations Density 15.656%

    No Known Activations