© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 2969
    Prev
    Next
    INDEX
    Explanations

    finally followed by punctuation or a new thought

    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
    ors
    -0.10
    è¹Ĭ
    -0.10
    éĻĶ
    -0.09
    osphere
    -0.09
     already
    -0.09
    ening
    -0.09
    åıįæŃ£
    -0.09
    ched
    -0.09
    olla
    -0.09
    é¦ĸåħĪ
    -0.09
    POSITIVE LOGITS
    æľīä¸Ģ天
    0.12
    born
    0.12
    egan
    0.10
     ÙĪÙĦÙĬس
    0.09
    ä¸ĸçķĮä¸ĬæľĢ
    0.09
     оказÑĭва
    0.09
    -reaching
    0.09
    cial
    0.09
    izer
    0.09
    CODE
    0.09
    Activations Density 0.009%

    No Known Activations