© 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. 22362
    Prev
    Next
    INDEX
    Explanations

    can or cannot ability

    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
    èĢĮä¸įæĺ¯
    -0.12
    贯穿
    -0.09
    Ñĩки
    -0.09
    ables
    -0.09
    iclass
    -0.08
    achi
    -0.08
    å¤Ħå¤Ħ
    -0.08
     Ole
    -0.08
     Ranch
    -0.08
    èĢĮæĺ¯
    -0.08
    POSITIVE LOGITS
    æľīä»»ä½ķ
    0.12
    éĢ¸
    0.11
    ä¸įåı¯èĥ½
    0.11
    ä¹Łä¸įåı¯èĥ½
    0.10
    æĢİä¹Īåı¯èĥ½
    0.10
    éķ£
    0.10
    毯
    0.10
    稽
    0.10
    çij¾
    0.10
    æĸ¹æīį
    0.10
    Activations Density 0.065%

    No Known Activations