Upload folder using huggingface_hub

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: google/gemma-2-2b-it
+base_model: google/gemma-2-2b-jpn-it
 language:
 - ja
 library_name: transformers
@@ -17,7 +17,7 @@ extra_gated_button_content: Acknowledge license
 
 # mlx-community/gemma-2-2b-jpn-it-8bit
 
-The Model [mlx-community/gemma-2-2b-jpn-it-8bit](https://huggingface.co/mlx-community/gemma-2-2b-jpn-it-8bit) was converted to MLX format from [google/gemma-2-2b-jpn-it](https://huggingface.co/google/gemma-2-2b-jpn-it) using mlx-lm version **0.17.1**.
+The Model [mlx-community/gemma-2-2b-jpn-it-8bit](https://huggingface.co/mlx-community/gemma-2-2b-jpn-it-8bit) was converted to MLX format from [google/gemma-2-2b-jpn-it](https://huggingface.co/google/gemma-2-2b-jpn-it) using mlx-lm version **0.18.2**.
 
 ## Use with mlx
 
@@ -29,5 +29,14 @@ pip install mlx-lm
 from mlx_lm import load, generate
 
 model, tokenizer = load("mlx-community/gemma-2-2b-jpn-it-8bit")
-response = generate(model, tokenizer, prompt="hello", verbose=True)
+
+prompt="hello"
+
+if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
+    messages = [{"role": "user", "content": prompt}]
+    prompt = tokenizer.apply_chat_template(
+        messages, tokenize=False, add_generation_prompt=True
+    )
+
+response = generate(model, tokenizer, prompt=prompt, verbose=True)
 ```

config.json

@@ -25,6 +25,10 @@
         "group_size": 64,
         "bits": 8
     },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 8
+    },
     "query_pre_attn_scalar": 224,
     "rms_norm_eps": 1e-06,
     "rope_theta": 10000.0,