© 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. 37945
    Prev
    Next
    INDEX
    Explanations

    severing or cutting

    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
    éªĪ
    -0.09
    ло
    -0.08
    å©Ĩ
    -0.08
    system
    -0.08
    ãģĹãĤĩãģĨ
    -0.08
    åĦ¡
    -0.08
    ä½ľèĢħæľ¬äºº
    -0.08
    abor
    -0.07
    عد
    -0.07
     Gecko
    -0.07
    POSITIVE LOGITS
    lessly
    0.11
    alk
    0.10
    sth
    0.09
    æľķ
    0.09
    wise
    0.09
     оÑĤд
    0.09
    ãĥªãĥ³ãĤ°
    0.09
    etting
    0.09
    ones
    0.09
    oning
    0.08
    Activations Density 0.013%

    No Known Activations