© 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. 37692
    Prev
    Next
    INDEX
    Explanations

    whopping, hefty amounts

    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
    estre
    -0.10
    ÙĨص
    -0.09
    AREST
    -0.09
    chè
    -0.09
    ensity
    -0.08
     повÑĭÑĪен
    -0.08
     obliv
    -0.08
    Opens
    -0.08
    ä¸ĵ人
    -0.08
    è°ħ
    -0.08
    POSITIVE LOGITS
    ä¹İ
    0.13
    lic
    0.10
    ly
    0.10
    çĦ¶æĺ¯
    0.10
    æ¦ľ
    0.09
    ãģĬãĤĬ
    0.09
    ingly
    0.08
    lij
    0.08
    yyy
    0.08
    çİĸ
    0.08
    Activations Density 0.004%

    No Known Activations