© 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. 37251
    Prev
    Next
    INDEX
    Explanations

    combines, joint, fusion, produce

    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
    ä¸İæѤ
    -0.10
    èĬĬ
    -0.08
    ãģ®ãģĬ
    -0.08
    озв
    -0.08
    æµı
    -0.08
     davran
    -0.08
    æıIJæĭĶ
    -0.08
     Fallon
    -0.08
    伺
    -0.08
    fox
    -0.08
    POSITIVE LOGITS
    ç»ĦåIJĪ
    0.16
    èģĶåIJĪ
    0.15
    íķ©
    0.15
    (comb
    0.14
     comb
    0.13
     joint
    0.13
    æ··åIJĪ
    0.13
    åıĮæĸ¹
    0.13
    åIJĪåĬĽ
    0.12
    comb
    0.12
    Activations Density 0.320%

    No Known Activations