© 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. 21189
    Prev
    Next
    INDEX
    Explanations

    whenever

    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
    arine
    -0.10
    åħļæĶ¯éĥ¨
    -0.09
    çĿ«
    -0.08
    race
    -0.08
    ÙĪØ·
    -0.08
    èĥ¥
    -0.08
    .Requires
    -0.08
    odore
    -0.08
    y
    -0.08
     Bund
    -0.08
    POSITIVE LOGITS
    æľīä»»ä½ķ
    0.10
    theless
    0.10
    ä»»ä½ķæŶåĢĻ
    0.10
    çļĦåį°è±¡
    0.10
    åĵªæĢķ
    0.09
    fy
    0.09
    abouts
    0.09
    æĪijè¿ĺæĺ¯
    0.09
    THING
    0.09
    dal
    0.09
    Activations Density 0.014%

    No Known Activations