Improve language tag

README.md

@@ -1,175 +1,189 @@
----
-base_model:
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-- Qwen/Qwen2.5-1.5B-Instruct
-tags:
-- merge
-- mergekit
-- lazymergekit
-- Qwen/Qwen2.5-1.5B-Instruct
----
-
-# Qwen2.5-2B-Instruct
-
-Qwen2.5-2B-Instruct is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
-
-## 🧩 Configuration
-
-```yaml
-dtype: bfloat16
-merge_method: passthrough
-slices:
-- sources:
-  - layer_range: [0, 2]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [1, 3]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [2, 4]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [3, 5]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [4, 6]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [5, 7]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [6, 8]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [7, 9]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [8, 10]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [9, 11]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [10, 12]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [11, 13]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [12, 14]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [13, 15]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [14, 16]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [16, 18]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [17, 19]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [18, 20]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [19, 21]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [20, 22]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [21, 23]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [22, 24]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [23, 25]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [24, 26]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [25, 27]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-- sources:
-  - layer_range: [26, 28]
-    model: Qwen/Qwen2.5-1.5B-Instruct
-```
-
-## 💻 Usage
-
-```python
-!pip install -qU transformers accelerate
-
-from transformers import AutoTokenizer
-import transformers
-import torch
-
-model = "win10/Qwen2.5-2B-Instruct"
-messages = [{"role": "user", "content": "What is a large language model?"}]
-
-tokenizer = AutoTokenizer.from_pretrained(model)
-prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-pipeline = transformers.pipeline(
-    "text-generation",
-    model=model,
-    torch_dtype=torch.float16,
-    device_map="auto",
-)
-
-outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
-print(outputs[0]["generated_text"])
+---
+base_model:
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-1.5B-Instruct
+tags:
+- merge
+- mergekit
+- lazymergekit
+- Qwen/Qwen2.5-1.5B-Instruct
+language:
+- zho
+- eng
+- fra
+- spa
+- por
+- deu
+- ita
+- rus
+- jpn
+- kor
+- vie
+- tha
+- ara
+---
+
+# Qwen2.5-2B-Instruct
+
+Qwen2.5-2B-Instruct is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+* [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct)
+
+## 🧩 Configuration
+
+```yaml
+dtype: bfloat16
+merge_method: passthrough
+slices:
+- sources:
+  - layer_range: [0, 2]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [1, 3]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [2, 4]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [3, 5]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [4, 6]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [5, 7]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [6, 8]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [7, 9]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [8, 10]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [9, 11]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [10, 12]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [11, 13]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [12, 14]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [13, 15]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [14, 16]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [16, 18]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [17, 19]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [18, 20]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [19, 21]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [20, 22]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [21, 23]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [22, 24]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [23, 25]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [24, 26]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [25, 27]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+- sources:
+  - layer_range: [26, 28]
+    model: Qwen/Qwen2.5-1.5B-Instruct
+```
+
+## 💻 Usage
+
+```python
+!pip install -qU transformers accelerate
+
+from transformers import AutoTokenizer
+import transformers
+import torch
+
+model = "win10/Qwen2.5-2B-Instruct"
+messages = [{"role": "user", "content": "What is a large language model?"}]
+
+tokenizer = AutoTokenizer.from_pretrained(model)
+prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+pipeline = transformers.pipeline(
+    "text-generation",
+    model=model,
+    torch_dtype=torch.float16,
+    device_map="auto",
+)
+
+outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
+print(outputs[0]["generated_text"])
 ```