cmedina/agenx-lora-training
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: use already-quantized 4-bit model for test

Browse Source
This commit is contained in:
Christian Medina
2026-07-02 12:09:49 -04:00
parent e3ea60e6c6
commit 3e26d7d37e
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
test_model_loading.py
View File

@@ -31,9 +31,9 @@ def test_model_loading():
print(" Loading model...") print(" Loading model...")
model = AutoModelForCausalLM.from_pretrained( model = AutoModelForCausalLM.from_pretrained(
"/data/models/Ornith-1.0-35B", "/data/models/Ornith-1.0-35B-4bit", # Use already-quantized 4-bit model
quantization_config=bnb_config,
device_map="auto", device_map="auto",
torch_dtype=torch.float16,
trust_remote_code=True, trust_remote_code=True,
low_cpu_mem_usage=True, low_cpu_mem_usage=True,
) )