riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-07-31 10:12:23 +06:00
Code Issues Actions 23 Packages Projects Releases Wiki Activity

Changing the test model in Quanto kv cache (#36670)

Browse Source 
changing model
This commit is contained in:
Mohamed Mekkouri 2025-03-13 12:23:34 +01:00 committed by GitHub
parent bc3d5781e7
commit 47cc4da351
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
tests/quantization/quanto_integration/test_quanto.py
View File

@ -448,17 +448,19 @@ class QuantoKVCacheQuantizationTest(unittest.TestCase):
@require_read_token
def test_quantized_cache(self):
EXPECTED_TEXT_COMPLETION = [
"Simply put, the theory of relativity states that 1) the speed of light is the same for all observers, and 2) the laws of physics are the same for all observers.\nThe first part of the theory is the most",
"My favorite all time favorite condiment is ketchup. I love it on everything. I love it on my eggs, my fries, my chicken, my burgers, my hot dogs, my sandwiches, my salads, my p",
"Simply put, the theory of relativity states that 1) time and space are not absolute, but are relative to the observer, and 2) the laws of physics are the same everywhere in the universe. This means that the speed of light is",
"My favorite all time favorite condiment is ketchup. I love how it adds a sweet and tangy flavor to my food. I also enjoy using it as a dip for fries, burgers, and grilled meats. It's a classic condiment that never",
]
prompts = [
"Simply put, the theory of relativity states that ",
"My favorite all time favorite condiment is ketchup.",
]
tokenizer = LlamaTokenizer.from_pretrained("meta-llama/Llama-2-7b-hf", pad_token="</s>", padding_side="left")
tokenizer = LlamaTokenizer.from_pretrained(
"unsloth/Llama-3.2-1B-Instruct", pad_token="</s>", padding_side="left"
)
model = LlamaForCausalLM.from_pretrained(
"meta-llama/Llama-2-7b-hf", device_map="sequential", torch_dtype=torch.float16
"unsloth/Llama-3.2-1B-Instruct", device_map="sequential", torch_dtype=torch.float16
)
inputs = tokenizer(prompts, return_tensors="pt", padding=True).to(torch_device)