© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 56563
    Prev
    Next
    INDEX
    Explanations

    of [number]

    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
    otas
    -0.10
    yne
    -0.08
    è·Ŀ
    -0.08
    éļĶå£ģ
    -0.08
    qh
    -0.08
    tte
    -0.08
    wen
    -0.08
    è¯·ä½ł
    -0.08
    LOOP
    -0.08
    ouses
    -0.08
    POSITIVE LOGITS
    诸å¤ļ
    0.17
    ä¸Ģç³»åĪĹ
    0.16
    ä¼Ĺå¤ļ
    0.15
    äºĨä¸Ģç³»åĪĹ
    0.12
    æĹłæķ°
    0.12
     many
    0.11
     многиÑħ
    0.11
    许å¤ļ
    0.11
     several
    0.11
    èĭ¥å¹²
    0.11
    Activations Density 0.043%

    No Known Activations