© 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. 9368
    Prev
    Next
    INDEX
    Explanations

    review contexts

    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
    çĦ¶æĺ¯
    -0.16
    CHA
    -0.10
    ities
    -0.10
    ¼
    -0.09
    itude
    -0.09
    rowse
    -0.09
    ä¸Ģå®ļçļĦ
    -0.09
    ircuit
    -0.09
    alog
    -0.09
    chaft
    -0.09
    POSITIVE LOGITS
    able
    0.14
    tube
    0.12
    ably
    0.11
    é¹Ĭ
    0.10
    çĥ¬
    0.10
    รม
    0.10
    aries
    0.09
    sic
    0.09
    èµ·è¯ī
    0.09
    è¾ĸ
    0.09
    Activations Density 0.047%

    No Known Activations