© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36609
    Prev
    Next
    INDEX
    Explanations

    common types

    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
    ä¸Ģ声
    -0.09
    furt
    -0.08
    anza
    -0.08
     Strap
    -0.08
    usat
    -0.08
    ãĤĮãģªãģĦ
    -0.08
    andin
    -0.08
    irmed
    -0.08
     Downing
    -0.08
    深深çļĦ
    -0.08
    POSITIVE LOGITS
    常è§ģçļĦ
    0.12
    Types
    0.11
    ç§įç±»
    0.11
    çļĦåŁºæľ¬
    0.11
     common
    0.11
    å¦Ĥä½ķåİ»
    0.11
     commonly
    0.11
     types
    0.10
    é«ĺåľ°
    0.10
    atypes
    0.10
    Activations Density 0.125%

    No Known Activations