© 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. 36545
    Prev
    Next
    INDEX
    Explanations

    ahead of

    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
    æ¤Ĵ
    -0.09
    为主é¢ĺçļĦ
    -0.09
    äº²çľ¼
    -0.09
     Middleton
    -0.09
    clist
    -0.09
    çĪ¬ä¸Ĭ
    -0.09
    ials
    -0.09
     ÑĤоÑĩно
    -0.09
    çŁį
    -0.08
    unal
    -0.08
    POSITIVE LOGITS
     ahead
    0.21
    é¢ĨåħĪ
    0.18
    ahead
    0.18
     Ahead
    0.17
    Ahead
    0.15
    æıIJåīį
    0.15
    åħĪè¡Į
    0.14
    æĬ¢åħĪ
    0.13
    -ahead
    0.13
    åīįéĿ¢
    0.13
    Activations Density 0.043%

    No Known Activations