© 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. 6669
    Prev
    Next
    INDEX
    Explanations

    adverbs followed by positive/negative adjectives

    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
    paren
    -0.09
    æĺ¾çĦ¶
    -0.09
    å»·
    -0.08
    åıł
    -0.08
    å¹ķ
    -0.08
    ä¸ļ
    -0.08
    AGR
    -0.08
    å¿ħå®ļ
    -0.08
     slightest
    -0.08
    æ¯ĭ
    -0.08
    POSITIVE LOGITS
    è¦ģåģļåĪ°
    0.10
    å¤ļåľ°
    0.10
    èĥ½åģļåĪ°
    0.10
    æľīäºĽ
    0.10
    ily
    0.09
    æĺ¯ä¸ª
    0.09
    ably
    0.09
    ä¸įè§ģ
    0.09
    æľīäºĽäºº
    0.09
    åīľ
    0.09
    Activations Density 0.033%

    No Known Activations