SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Fix : add require_read_token for gemma2 gated model (#35687)

Browse Source 
fix gemma2 gated model test
This commit is contained in:
Mohamed Mekkouri
2025-01-14 11:47:05 +01:00
committed by GitHub
parent df2a812e95
commit a11041ffad
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/quantization/ggml/test_ggml.py
View File

@@ -18,6 +18,7 @@ import unittest
from transformers import AddedToken, AutoModelForCausalLM, AutoModelForSeq2SeqLM, AutoTokenizer from transformers import AddedToken, AutoModelForCausalLM, AutoModelForSeq2SeqLM, AutoTokenizer
from transformers.testing_utils import ( from transformers.testing_utils import (
require_gguf, require_gguf,
require_read_token,
require_torch_gpu, require_torch_gpu,
slow, slow,
torch_device, torch_device,
@@ -880,6 +881,7 @@ class GgufIntegrationTests(unittest.TestCase):
EXPECTED_TEXT = "Hello! 👋\n\nI'm a large language model" EXPECTED_TEXT = "Hello! 👋\n\nI'm a large language model"
self.assertEqual(tokenizer.decode(out[0], skip_special_tokens=True), EXPECTED_TEXT) self.assertEqual(tokenizer.decode(out[0], skip_special_tokens=True), EXPECTED_TEXT)
@require_read_token
def test_gemma2_weights_conversion_fp32(self): def test_gemma2_weights_conversion_fp32(self):
original_model = AutoModelForCausalLM.from_pretrained( original_model = AutoModelForCausalLM.from_pretrained(
self.original_gemma2_model_id, self.original_gemma2_model_id,