© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7489
    Prev
    Next
    INDEX
    Explanations

    concept or idea

    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
    山寨
    -0.09
    å¾ģ
    -0.08
    è¿ĻæŶ
    -0.08
    itative
    -0.08
     crafts
    -0.08
    çľł
    -0.08
    ype
    -0.08
    çĹĩ
    -0.08
    ature
    -0.08
     oscill
    -0.08
    POSITIVE LOGITS
     concept
    0.22
     idea
    0.22
    çļĦæ¦Ĥ念
    0.21
    æ¦Ĥ念
    0.21
    idea
    0.19
     concepts
    0.19
    concept
    0.19
    æĥ³æ³ķ
    0.18
    çļĦçIJĨ念
    0.17
    çIJĨ念
    0.17
    Activations Density 0.075%

    No Known Activations