© 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. 37046
    Prev
    Next
    INDEX
    Explanations

    separation or independence

    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
    _Inter
    -0.10
    ument
    -0.09
    pu
    -0.09
    çĬ¹
    -0.09
    æŶä¸į
    -0.09
    encing
    -0.08
     никÑĤо
    -0.08
    .parametrize
    -0.08
    æ½Ń
    -0.08
    веден
    -0.08
    POSITIVE LOGITS
    离å¼Ģ
    0.11
    天çĮ«
    0.11
    åıĽ
    0.11
    èĦ±ç¦»
    0.10
    åĪĨ离
    0.10
    canf
    0.10
    éĽ¢éĸĭ
    0.10
    åĽ½éĻħåIJĪä½ľ
    0.09
    éĹ¹
    0.09
    ç¿Ĭ
    0.09
    Activations Density 0.075%

    No Known Activations