© 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. 58308
    Prev
    Next
    INDEX
    Explanations

    states and territories except

    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
    ansa
    -0.10
    ansas
    -0.10
    æī¹
    -0.09
    å¤ļ家
    -0.09
    reeze
    -0.09
    å¾Ĺ天
    -0.09
    åĬŁåĬĽ
    -0.09
    å¿ĥä¸ŃçļĦ
    -0.09
    å¤ļä¸ªåĽ½å®¶
    -0.09
    æľªå¿ħ
    -0.09
    POSITIVE LOGITS
     except
    0.14
    except
    0.13
    éĥ½èĥ½
    0.12
    éĥ½åı¯ä»¥
    0.11
    éĥ½æľī
    0.11
    éĥ½è¦ģ
    0.10
    3
    0.10
     Except
    0.10
    1
    0.10
    8
    0.10
    Activations Density 0.108%

    No Known Activations