© 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. 7621
    Prev
    Next
    INDEX
    Explanations

    probably followed by common words

    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
    -0.12
    urance
    -0.11
    fty
    -0.10
    nit
    -0.10
    å±ħçĦ¶
    -0.09
    ç«ŁçĦ¶
    -0.09
    ãģĿãĤĤãģĿãĤĤ
    -0.09
    manship
    -0.09
    忽çĦ¶
    -0.09
    éĥ¨åĪĨåľ°åĮº
    -0.09
    POSITIVE LOGITS
    mente
    0.13
     (?)
    0.10
    æĦıä¹īä¸ĬçļĦ
    0.10
    以ä¸ĭåĩłä¸ª
    0.09
    ity
    0.09
    hood
    0.09
    ment
    0.09
    idente
    0.09
    ERING
    0.09
    izer
    0.09
    Activations Density 0.016%

    No Known Activations