© 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. 51196
    Prev
    Next
    INDEX
    Explanations

    together, combined, with each other

    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
     __________________↵↵
    -0.09
    æĴ¥
    -0.09
    ibel
    -0.08
    etur
    -0.08
    essler
    -0.08
     Cra
    -0.08
    å¸ĪäºĭåĬ¡æīĢ
    -0.08
    runner
    -0.08
    eneg
    -0.08
    å´ŀ
    -0.08
    POSITIVE LOGITS
     together
    0.28
    ä¸Ģèµ·
    0.26
    åħ±åIJĮ
    0.26
     Together
    0.24
    åħ±åIJĮåĬªåĬĽ
    0.24
    Together
    0.22
    åľ¨ä¸Ģèµ·
    0.21
     zusammen
    0.20
     вмеÑģÑĤе
    0.20
    ä¸Ģèµ·æĿ¥
    0.19
    Activations Density 0.051%

    No Known Activations