Upload folder using huggingface_hub

README.md

@@ -0,0 +1,68 @@
+---
+base_model:
+- ce-lery/dolly-japanese-gpt-1b-clone
+- rinna/japanese-gpt-1b
+tags:
+- merge
+- mergekit
+- lazymergekit
+- ce-lery/dolly-japanese-gpt-1b-clone
+- rinna/japanese-gpt-1b
+---
+
+# jp-gpt-ab-dareties
+
+jp-gpt-ab-dareties is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
+* [ce-lery/dolly-japanese-gpt-1b-clone](https://huggingface.co/ce-lery/dolly-japanese-gpt-1b-clone)
+* [rinna/japanese-gpt-1b](https://huggingface.co/rinna/japanese-gpt-1b)
+
+## 🧩 Configuration
+
+```yaml
+slices:
+- sources:
+  - layer_range: [0, 24]
+    model: ce-lery/dolly-japanese-gpt-1b-clone
+    parameters:
+      density: [1, 0.7, 0.1]
+      weight: 1.0
+  - layer_range: [0, 24]
+    model: rinna/japanese-gpt-1b
+    parameters:
+      density: 0.33
+      weight:
+        - filter: mlp
+          value: 0.5
+        - value: 0
+merge_method: dare_ties
+base_model: aipib/Tinyllama-moe3
+parameters:
+  normalize: true
+  int8_mask: true
+dtype: bfloat16
+```
+
+## 💻 Usage
+
+```python
+!pip install -qU transformers accelerate
+
+from transformers import AutoTokenizer
+import transformers
+import torch
+
+model = "aipib/jp-gpt-ab-dareties"
+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"])
+```