© 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. 34664
    Prev
    Next
    INDEX
    Explanations

    After

    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
    æłı
    -0.10
    èĨĤ
    -0.10
    æ§Ľ
    -0.09
    å½ĺ
    -0.09
    å¹³æĿ¿
    -0.08
    è¿Ļåıªæĺ¯
    -0.08
    åĩłåĪĨ
    -0.08
    éĤ£æŶåĢĻ
    -0.08
    nist
    -0.08
    æķıæį·
    -0.08
    POSITIVE LOGITS
    æĹłå¥Ī
    0.11
    ä»Ķç»Ĩ
    0.10
    è¿Ļä¹Īå¤ļå¹´
    0.10
    笺
    0.10
    è½½
    0.10
    èĢĥèĻijåĪ°
    0.10
    ç©·
    0.10
    æĿ¥çľĭ
    0.10
    5
    0.09
    çľĭçľĭ
    0.09
    Activations Density 0.047%

    No Known Activations