© 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. 30956
    Prev
    Next
    INDEX
    Explanations

    straight followed by time or team

    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
    åħ¶å®ĥéĹ®é¢ĺ
    -0.10
    struction
    -0.09
    agua
    -0.09
    éĻIJãĤĬ
    -0.09
     slight
    -0.09
    æľ
    -0.08
    rtc
    -0.08
    éħĬ
    -0.08
    Ø´
    -0.08
    妾
    -0.08
    POSITIVE LOGITS
    lining
    0.14
    pped
    0.14
    ly
    0.13
    liner
    0.13
    zeitig
    0.12
    ening
    0.12
    ness
    0.11
    NESS
    0.11
    gün
    0.10
    away
    0.10
    Activations Density 0.010%

    No Known Activations