© 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. 63994
    Prev
    Next
    INDEX
    Explanations

    inen

    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
    Whole
    -0.28
    æľī幸
    -0.24
     drafts
    -0.24
     bonded
    -0.24
    æĸ°æ¬¾
    -0.23
     enjo
    -0.23
    rush
    -0.23
    Russian
    -0.23
     Supplementary
    -0.23
    Ear
    -0.23
    POSITIVE LOGITS
    æĬ¥ä»·
    0.27
    uib
    0.27
    åīįæĻ¯
    0.26
    lein
    0.26
    çļĦè¯Ŀé¢ĺ
    0.26
    æŁıæŀĹ
    0.26
    åIJ»
    0.26
    å¥Ĺ
    0.26
    æĸ¹éĿ¢çļĦ
    0.26
    å´İ
    0.25
    Activations Density 23.519%

    No Known Activations