© 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. 52091
    Prev
    Next
    INDEX
    Explanations

    idiom, adjective, suffix -ly

    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
    è¿Ļ对
    -0.09
    adora
    -0.09
    ouce
    -0.09
    ifle
    -0.09
    幺
    -0.09
    parm
    -0.09
    åĽłä¸ºåľ¨
    -0.08
    INO
    -0.08
    _backward
    -0.08
    _combine
    -0.08
    POSITIVE LOGITS
    ical
    0.13
    éĢļãĤĬ
    0.12
    ä¸įæ¸ħ
    0.11
    ues
    0.10
    大åħ¨
    0.10
    家çĶ¨
    0.10
    gle
    0.09
    ary
    0.09
    awe
    0.09
    ously
    0.09
    Activations Density 0.094%

    No Known Activations