Upload folder using huggingface_hub

README.md

@@ -0,0 +1,65 @@
+---
+base_model:
+- NousResearch/Hermes-2-Pro-Mistral-7B
+- mistralai/Mistral-7B-Instruct-v0.2
+tags:
+- merge
+- mergekit
+- lazymergekit
+- NousResearch/Hermes-2-Pro-Mistral-7B
+- mistralai/Mistral-7B-Instruct-v0.2
+---
+
+# TranscriptAnalyzer-7B
+
+TranscriptAnalyzer-7B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
+* [NousResearch/Hermes-2-Pro-Mistral-7B](https://huggingface.co/NousResearch/Hermes-2-Pro-Mistral-7B)
+* [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
+
+## 🧩 Configuration
+
+```yaml
+base_model: mistralai/Mistral-7B-v0.1
+dtype: float16
+merge_method: dare_ties
+parameters:
+  density: 0.6  # Optimal pour 2 modèles
+
+slices:
+  - sources:
+    - model: NousResearch/Hermes-2-Pro-Mistral-7B
+      layer_range: [0, 32]
+      parameters:
+        weight: 0.65  # 65% Hermes (analyse)
+        density: 0.7
+    - model: mistralai/Mistral-7B-Instruct-v0.2
+      layer_range: [0, 32]
+      parameters:
+        weight: 0.35  # 35% Mistral (rapidité)
+        density: 0.6
+```
+
+## 💻 Usage
+
+```python
+!pip install -qU transformers accelerate
+
+from transformers import AutoTokenizer
+import transformers
+import torch
+
+model = "alantcoding/TranscriptAnalyzer-7B"
+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"])
+```