diff --git a/server/text_generation_server/utils/gptq/quant_linear.py b/server/text_generation_server/utils/gptq/quant_linear.py index 00f902ce..6e939115 100644 --- a/server/text_generation_server/utils/gptq/quant_linear.py +++ b/server/text_generation_server/utils/gptq/quant_linear.py @@ -257,7 +257,7 @@ try: c_mask = (offs_am[:, None] < M) & (offs_bk[None, :] < K) tl.store(c_ptrs, accumulator, mask=c_mask) except: - print('trioton not installed.') + print('triton not installed.') def matmul248(input, qweight, scales, qzeros, g_idx, bits, maxq):