© 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. 8468
    Prev
    Next
    INDEX
    Explanations

    correctness or incorrectness

    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
    lan
    -0.10
    ary
    -0.09
    çī¹æľĹ
    -0.09
    chat
    -0.09
     lan
    -0.09
    Ñıн
    -0.09
    åĽºåĮĸ
    -0.08
    éĤĪ
    -0.08
    forth
    -0.08
    .intellij
    -0.08
    POSITIVE LOGITS
    eous
    0.14
    /right
    0.12
    姿åĬ¿
    0.11
    管è¾ĸ
    0.10
     functioning
    0.10
    -sizing
    0.09
    yntax
    0.09
    âĸ¸
    0.09
    é£İæļ´
    0.09
     kullanımı
    0.08
    Activations Density 0.098%

    No Known Activations