© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 31501
    Prev
    Next
    INDEX
    Explanations

    offers, promotions, and rewards

    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
    主å¸Ń
    -0.09
    ØŃÙģظ
    -0.09
    åİĭå®ŀ
    -0.08
    ä¸ĢåįĬ
    -0.08
    åģļçļĦäºĭ
    -0.08
    驳
    -0.08
    ensored
    -0.08
    ritel
    -0.08
     CNS
    -0.08
    èĦŀ
    -0.08
    POSITIVE LOGITS
    å¥ĸåĬ±
    0.14
    ðŁİģ
    0.11
    ä¿ĥéĶĢ
    0.11
    ä¼ĺæĥł
    0.10
     rewards
    0.10
    _reward
    0.10
    Reward
    0.10
    ãĤŃãĥ£ãĥ³ãĥļãĥ¼ãĥ³
    0.10
     Rewards
    0.10
    æ´»åĬ¨
    0.10
    Activations Density 0.091%

    No Known Activations