© 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. 20749
    Prev
    Next
    INDEX
    Explanations

    reading and readers

    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
    /from
    -0.11
    forth
    -0.10
     readonly
    -0.10
    antar
    -0.09
    ities
    -0.09
    çĸ¹
    -0.09
    ally
    -0.08
    emente
    -0.08
     crib
    -0.08
    ement
    -0.08
    POSITIVE LOGITS
    just
    0.16
    iness
    0.13
    ings
    0.12
    /view
    0.12
    vertise
    0.12
    ä¸įåĩºæĿ¥
    0.12
    é¢Ĩä¼ļ
    0.12
    mission
    0.11
     aloud
    0.11
    vertiser
    0.11
    Activations Density 0.076%

    No Known Activations