© 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. 7072
    Prev
    Next
    INDEX
    Explanations

    Mumbai, Bombay

    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
    úa
    -0.11
    ural
    -0.11
    tones
    -0.10
    ibri
    -0.09
    umont
    -0.09
    ibs
    -0.09
    quo
    -0.09
    夫
    -0.09
    hear
    -0.08
    bred
    -0.08
    POSITIVE LOGITS
    åŃĶéĽĢ
    0.11
    perator
    0.10
    å´ĸ
    0.10
    滩
    0.09
    ¨
    0.09
    ï¸
    0.09
    å²Ļ
    0.08
    erset
    0.08
    æľĢåıĹ欢è¿İ
    0.08
     sophistic
    0.08
    Activations Density 0.006%

    No Known Activations