© 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. 6730
    Prev
    Next
    INDEX
    Explanations

    eyes closed or open

    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
    éģı
    -0.09
    é¥Ń
    -0.09
    å¤ĩ
    -0.09
    ç©¿çĿĢ
    -0.08
    åĪĢ
    -0.08
    azzo
    -0.08
    .ls
    -0.08
     Massive
    -0.08
    éĩį
    -0.08
    æĪĸå¤ļ
    -0.08
    POSITIVE LOGITS
    éĹŃ
    0.16
    çľ¯
    0.14
    çĿģå¼Ģ
    0.13
    çľ¼çĿĽ
    0.13
    çĿģ
    0.13
    åħ³éĹŃ
    0.13
    åıĮçľ¼
    0.13
     closed
    0.13
    éĸī
    0.12
     eyes
    0.12
    Activations Density 0.037%

    No Known Activations