© 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. 48861
    Prev
    Next
    INDEX
    Explanations

    allows, required, expects

    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
    utory
    -0.10
    æ²Ļ
    -0.08
    estr
    -0.08
    åį´è¢«
    -0.08
    uilder
    -0.08
    éĵŃ
    -0.08
    ufe
    -0.08
     Nik
    -0.08
    éħ¥
    -0.07
    å½ĵæĪIJ
    -0.07
    POSITIVE LOGITS
    ä¸įåħģ许
    0.14
    è¦ģæ±Ĥ
    0.14
     ìļĶ구
    0.12
    æĩĭ
    0.11
     позволÑı
    0.11
    æľīè¿Ļæł·çļĦ
    0.11
    è¿Ļä¹Īåģļ
    0.11
     demands
    0.11
    è¿Ļæł·åģļ
    0.10
     necess
    0.10
    Activations Density 0.092%

    No Known Activations