© 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. 18067
    Prev
    Next
    INDEX
    Explanations

    difference of / between

    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
    æĪļ
    -0.10
    å¿ĥ缮
    -0.10
    è´¿
    -0.09
    ç¶ļãģį
    -0.08
    esis
    -0.08
    æ±ŁåĮĹ
    -0.08
    arya
    -0.08
    лег
    -0.08
    ãĤĴè¿İãģĪ
    -0.08
    ä¹Łæľīä¸ĢäºĽ
    -0.08
    POSITIVE LOGITS
    å·®è·Ŀ
    0.21
    å·®å¼Ĥ
    0.19
    çĽ¸å·®
    0.18
     difference
    0.18
    å·®
    0.18
    å¹ħ度
    0.17
    å·®åĪ«
    0.17
    difference
    0.16
    ifference
    0.14
    Difference
    0.14
    Activations Density 0.188%

    No Known Activations