© 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. 23-TRANSCODER-HP
    7. 24960
    Prev
    Next
    INDEX
    Explanations

    Italian

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ár
    -0.30
    chor
    -0.30
    amus
    -0.29
    otch
    -0.28
    åij¼åĴĮ
    -0.27
    át
    -0.26
    èIJ½åľ¨
    -0.26
    座è°Ī
    -0.25
    ática
    -0.25
    udder
    -0.25
    POSITIVE LOGITS
    æĦı大åĪ©
    0.54
     Italian
    0.54
     Italians
    0.49
     Italy
    0.48
    æĪIJéĥ½
    0.47
     italian
    0.47
    Italian
    0.46
    å®ľå®¾
    0.45
    UniversitÃł
    0.45
    æĪIJéĥ½å¸Ĥ
    0.44
    Activations Density 0.159%

    No Known Activations