© 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. 8498
    Prev
    Next
    INDEX
    Explanations

    inherent characteristics

    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
    mans
    -0.09
    缴æݥ影åĵį
    -0.09
    空åīį
    -0.08
    ook
    -0.08
    apon
    -0.08
    ano
    -0.08
    @m
    -0.08
    æ¿Ģ
    -0.08
    ä¸ĢæŶ
    -0.08
    ienza
    -0.08
    POSITIVE LOGITS
    ities
    0.12
    ìłģìľ¼ë¡ľ
    0.10
    ITIES
    0.10
    ä½ļ
    0.09
    à¦
    0.09
     Elm
    0.09
    cept
    0.08
    ding
    0.08
    ease
    0.08
    çĵ¦
    0.08
    Activations Density 0.021%

    No Known Activations