© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7369
    Prev
    Next
    INDEX
    Explanations

    put

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åľ¨åĮĹ京
    -0.09
    FI
    -0.08
    èĽĽ
    -0.08
    缩
    -0.08
    bst
    -0.08
    é©»
    -0.08
    ror
    -0.07
    èģĮåĬ¡
    -0.07
    osta
    -0.07
    æĢĿè·¯
    -0.07
    POSITIVE LOGITS
     into
    0.12
    StrLn
    0.11
     aside
    0.11
    è¿Ľåİ»
    0.10
     vÃło
    0.10
    åľ¨ä¸Ģèµ·
    0.10
    å±ıéļľ
    0.10
     forth
    0.10
    Into
    0.10
    ç¾ŀ
    0.10
    Activations Density 0.138%

    No Known Activations