© 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. 20579
    Prev
    Next
    INDEX
    Explanations

    in higher detail

    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
    èIJ¥
    -0.09
     distant
    -0.09
     ведÑĮ
    -0.09
    бой
    -0.09
    acky
    -0.08
    iously
    -0.08
    ç²¾éĢļ
    -0.08
    éķ¿çĽ¸
    -0.08
     Primitive
    -0.08
    _protocol
    -0.08
    POSITIVE LOGITS
    èĭ±è¯Ń
    0.10
    æµĵ
    0.09
    ä¼ļéķ¿
    0.09
    æ¸ħæĻ°
    0.09
    ç®Ģ
    0.09
    ä¸ķ
    0.09
    amel
    0.08
    åĪĨä¼ļ
    0.08
    书éĿ¢
    0.08
    鬣
    0.08
    Activations Density 0.099%

    No Known Activations