© 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. 2728
    Prev
    Next
    INDEX
    Explanations

    life choices and support

    np_acts-logits-general · gemini-2.5-flash-lite

    Sentences and phrases about personal growth, therapy, healing, and emotional support—self-help / counselling language focused on feelings, confidence, choices, and recovery.

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    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
    æĥħä¾£
    -0.09
    èĬĿ
    -0.08
    åĨ»ç»ĵ
    -0.08
    åĪĺæµ·
    -0.08
     ÑģÑĤÑĥд
    -0.08
    举æŃ¢
    -0.08
    åĪ¶æľį
    -0.08
    entious
    -0.08
    (Utils
    -0.08
    orny
    -0.08
    POSITIVE LOGITS
    æݥ纳
    0.09
    è¿ĺæĮº
    0.09
    çļĦèĥ½éĩı
    0.09
    å°ıå°ıçļĦ
    0.08
     ener
    0.08
     Raq
    0.08
    earned
    0.08
     Bren
    0.08
    ä¸įå¤ļ
    0.08
    æĮīä¸ĭ
    0.08
    Activations Density 0.346%

    No Known Activations