© 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. 8735
    Prev
    Next
    INDEX
    Explanations

    legal citation prefixes

    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
    éĿłè°±
    -0.10
    æijŃ
    -0.10
    ern
    -0.09
    fty
    -0.09
    eres
    -0.09
    å²ģ以ä¸ĭ
    -0.09
     overshadow
    -0.08
    aven
    -0.08
    æ§IJ
    -0.08
    atched
    -0.08
    POSITIVE LOGITS
    ä¸ĢæŶ
    0.10
    æĥº
    0.09
    å¾ģæ±Ĥ
    0.08
    åį³ä¾¿
    0.08
    vier
    0.08
    gratis
    0.08
     Lun
    0.08
    -piece
    0.08
    人士
    0.08
    quer
    0.08
    Activations Density 0.548%

    No Known Activations