© 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. 9680
    Prev
    Next
    INDEX
    Explanations

    B starters

    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
    odore
    -0.11
    åºŃéĻ¢
    -0.10
    ordan
    -0.09
    abelle
    -0.09
    adays
    -0.09
    éĻ¢
    -0.08
    ombie
    -0.08
    IMUM
    -0.08
    oner
    -0.08
    vet
    -0.08
    POSITIVE LOGITS
    ç³Ĭæ¶Ĥ
    0.09
    ÑĢаг
    0.09
    èİĺ
    0.09
    åĬłåĢį
    0.08
    çĬĢ
    0.08
    pig
    0.08
    å¨ģ
    0.08
    ĶĦ
    0.08
    éĥ½ä¸įä¼ļ
    0.08
    icom
    0.08
    Activations Density 0.093%

    No Known Activations