© 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. 21741
    Prev
    Next
    INDEX
    Explanations

    any noun

    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
     единÑģÑĤв
    -0.10
    ä¸Ģåı£æ°Ķ
    -0.09
    èįī
    -0.09
    åľ¨æĪijçļĦ
    -0.09
    å°º
    -0.09
    çĵľ
    -0.09
    çī§
    -0.08
     stom
    -0.08
    个åĪ«
    -0.08
    osphere
    -0.08
    POSITIVE LOGITS
    åıªè¦ģæĺ¯
    0.16
    åĩ¡æĺ¯
    0.15
    åıªè¦ģæľī
    0.12
    åıªè¦ģä½ł
    0.11
    åĩ¡
    0.11
    åıªè¦ģ
    0.10
     whenever
    0.10
    类似çļĦ
    0.10
    æ¶īåıĬ
    0.09
    éĤ£ç§į
    0.09
    Activations Density 0.101%

    No Known Activations