© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8125
    Prev
    Next
    INDEX
    Explanations

    para, meta, substituents, p-

    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
    èł¡
    -0.11
    вол
    -0.09
    çļĦæķĪæŀľ
    -0.09
    бан
    -0.08
    æĬ¿
    -0.08
     Auburn
    -0.08
    åīįè¿°
    -0.08
    çĶ·å¥³
    -0.08
    èĵ¬åĭĥ
    -0.08
    /licenses
    -0.08
    POSITIVE LOGITS
    chte
    0.10
    è¯Ĩ
    0.09
    holm
    0.09
    eden
    0.09
    anus
    0.09
    æĮĤ
    0.09
    穷人
    0.08
    èIJģ
    0.08
     Benchmark
    0.08
    variants
    0.08
    Activations Density 0.244%

    No Known Activations