© 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. 20135
    Prev
    Next
    INDEX
    Explanations

    recognize as

    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
    辫
    -0.09
    ROUGH
    -0.09
    seat
    -0.09
    iture
    -0.09
    оÑĤов
    -0.08
    溱
    -0.08
     McCabe
    -0.08
     Salisbury
    -0.08
    uters
    -0.08
    STANCE
    -0.08
    POSITIVE LOGITS
     recognize
    0.11
    认åı¯
    0.10
     Recogn
    0.10
    nowled
    0.10
    æĹĮ
    0.10
    认è¯ĨåĪ°
    0.10
    ond
    0.09
     recognizes
    0.09
     valid
    0.09
    recognized
    0.09
    Activations Density 0.086%

    No Known Activations