© 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. 37162
    Prev
    Next
    INDEX
    Explanations

    cleavage, clearing, Cleveland, Cleary

    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
    eros
    -0.11
    ooter
    -0.10
     xuyên
    -0.10
    rovers
    -0.09
    icity
    -0.09
    ãģĦãĤĭ
    -0.09
    manship
    -0.09
    sWith
    -0.08
    adero
    -0.08
    éĥĩ
    -0.08
    POSITIVE LOGITS
    ص
    0.12
    -eyed
    0.11
    èĪ·
    0.11
    lest
    0.10
    é«®
    0.10
    å¹²åĩĢ
    0.10
    å¤ł
    0.10
    lection
    0.10
    egt
    0.10
    rons
    0.09
    Activations Density 0.017%

    No Known Activations