© 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. 57715
    Prev
    Next
    INDEX
    Explanations

    making a case

    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
     Grat
    -0.10
    æºī
    -0.09
     Neville
    -0.08
    çºłæŃ£
    -0.08
     nad
    -0.08
    -original
    -0.08
    é«®
    -0.08
     Schwe
    -0.08
    icher
    -0.08
    æ¯Ķè¾ĥå¤ļ
    -0.08
    POSITIVE LOGITS
     Arguments
    0.11
    çIJĨçĶ±
    0.11
     against
    0.10
    Arguments
    0.10
     arguments
    0.09
    继ç»Ńä¿ĿæĮģ
    0.09
    å¹²èĦĨ
    0.09
    åıį对
    0.09
    继ç»Ń
    0.09
    _again
    0.09
    Activations Density 0.055%

    No Known Activations