© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52523
    Prev
    Next
    INDEX
    Explanations

    just followed by a past tense verb

    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
    äºĨåIJ§
    -0.11
    poke
    -0.10
    继ç»Ńä¿ĿæĮģ
    -0.09
    æ¯ıåij¨
    -0.09
    åħ·ä½ĵæĥħåĨµ
    -0.09
    缮åīįæŃ£åľ¨
    -0.09
    缮åīįå·²
    -0.09
    éĶ¢
    -0.09
    驹
    -0.09
    çī¢è®°
    -0.09
    POSITIVE LOGITS
    aneous
    0.09
     dens
    0.09
    åŃij
    0.08
    é¢ģåıij
    0.08
    èģĨ
    0.08
    odial
    0.08
    åıįè¿ĩæĿ¥
    0.08
    egt
    0.08
     dial
    0.08
    çī©
    0.08
    Activations Density 0.035%

    No Known Activations