From da4e0dc4ddf687a2c2667a40856d5e8f40f938af Mon Sep 17 00:00:00 2001 From: Rahul Tuli Date: Sun, 6 Oct 2024 21:50:30 +0000 Subject: [PATCH] Set Sparse compression to save_compressed --- .../transformers/sparsification/compressed_tensors_utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/llmcompressor/transformers/sparsification/compressed_tensors_utils.py b/src/llmcompressor/transformers/sparsification/compressed_tensors_utils.py index 4ce00b9c8..c02621cd2 100644 --- a/src/llmcompressor/transformers/sparsification/compressed_tensors_utils.py +++ b/src/llmcompressor/transformers/sparsification/compressed_tensors_utils.py @@ -95,7 +95,7 @@ def save_pretrained_wrapper( "skip_compression_stats=True" ) sparsity_config = SparsityConfigMetadata.from_pretrained( - model, state_dict=state_dict, compress=False + model, state_dict=state_dict, compress=save_compressed ) quantization_format = infer_quantization_format(