© 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. 21050
    Prev
    Next
    INDEX
    Explanations

    prefixes like 'Mc' followed by name suffixes

    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
    踯
    -0.10
    аÑĨион
    -0.09
     пÑĢид
    -0.09
    å®īå±ħ
    -0.08
    aben
    -0.08
    CADE
    -0.08
    ãģıãĤĮ
    -0.08
    ably
    -0.08
    Ñģов
    -0.08
    ÑģÑĮ
    -0.08
    POSITIVE LOGITS
    ras
    0.10
    ience
    0.10
    ened
    0.09
    ening
    0.09
    raw
    0.09
    ran
    0.08
    ra
    0.08
    ition
    0.08
    åħ·ä½ĵæĥħåĨµ
    0.08
    ega
    0.08
    Activations Density 0.022%

    No Known Activations