© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 9133
    Prev
    Next
    INDEX
    Explanations

    flip or switch actions

    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
    èı½
    -0.11
    æħ
    -0.10
    被æī§è¡Į
    -0.10
    ums
    -0.09
    emy
    -0.09
    嵯
    -0.09
    MENTS
    -0.09
    ç¥ĵ
    -0.09
    orns
    -0.09
    ä¸Ģè½®
    -0.09
    POSITIVE LOGITS
    ç¿»
    0.12
    è¿ĩæĿ¥
    0.10
     flip
    0.10
     Flip
    0.10
     switches
    0.10
    switch
    0.09
    æĬĽ
    0.09
    æĹ¶æĬ¥
    0.09
     switch
    0.09
     flipped
    0.09
    Activations Density 0.024%

    No Known Activations