© 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. 130767
    Prev
    Next
    INDEX
    Explanations

    Programming code snippets

    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
    éģĹ产
    -0.28
    åĨ¶
    -0.28
     bergen
    -0.25
    èµĦ
    -0.25
     зна
    -0.24
    æ¸Ĭ
    -0.24
    :
    -0.24
    容
    -0.24
    éī´
    -0.24
    ivity
    -0.24
    POSITIVE LOGITS
     RFC
    0.28
    аÑĨион
    0.26
    çļĦ身份
    0.25
    unbind
    0.25
    RFC
    0.25
     stale
    0.24
    å°±è¿ij
    0.24
    ä¸ĩåħ¬éĩĮ
    0.24
    preter
    0.24
     testData
    0.24
    Activations Density 0.666%

    No Known Activations