© 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. 52441
    Prev
    Next
    INDEX
    Explanations

    fifth, 9, 1

    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
    åĶ¾
    -0.09
    åįļä¼ļ
    -0.09
    æĹº
    -0.09
     ÐļÑĥ
    -0.09
    ä¸ĢåľĪ
    -0.09
    apg
    -0.08
    alten
    -0.08
    ä»İå°ı
    -0.08
    çģµ
    -0.08
    edBy
    -0.08
    POSITIVE LOGITS
    æ¾į
    0.09
    åĽ¾ä¹¦é¦Ĩ
    0.09
    æ¹ĸæ³Ĭ
    0.09
    joy
    0.09
    æĺ¶
    0.09
    çĢĽ
    0.08
    ror
    0.08
    бол
    0.08
    涡
    0.08
    å¯Įæ°ij
    0.08
    Activations Density 0.047%

    No Known Activations