© 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. 32281
    Prev
    Next
    INDEX
    Explanations

    successfully working, documented, effective

    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
    æĶ¾å®½
    -0.09
    ilon
    -0.09
    享
    -0.09
    è®®
    -0.09
    iae
    -0.09
    æĹ»
    -0.08
    çķĮ
    -0.08
    çĶ²
    -0.08
    交
    -0.08
    åħ³å¿ĥ
    -0.08
    POSITIVE LOGITS
    æķĪæŀľ
    0.20
     effectiveness
    0.18
    æĪIJæķĪ
    0.16
    çļĦæķĪæŀľ
    0.14
    è§ģæķĪ
    0.14
     íļ¨ê³¼
    0.14
    æľīæķĪæĢ§
    0.14
    æľīæķĪ
    0.13
     effective
    0.13
    æķĪåĬĽ
    0.13
    Activations Density 0.346%

    No Known Activations