Upload folder using huggingface_hub

README.md

@@ -1,58 +1,50 @@
 ---
-license: cc-by-nc-4.0
-tags:
-- merge
-- mergekit
-- lazymergekit
-- automerger
 base_model:
+- eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2
 - yam-peleg/Experiment27-7B
----
+library_name: transformers
+tags:
+- mergekit
+- merge
 
-# OgnoExperiment27-7B
+---
+# merge
 
-OgnoExperiment27-7B is an automated merge created by [Maxime Labonne](https://huggingface.co/mlabonne) using the following configuration.
-* [yam-peleg/Experiment27-7B](https://huggingface.co/yam-peleg/Experiment27-7B)
+This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
 
-## 🧩 Configuration
+## Merge Details
+### Merge Method
 
-```yaml
-models:
-- model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO
-  # No parameters necessary for base model
-- model: yam-peleg/Experiment27-7B
-  parameters:
-    density: 0.53
-    weight: 0.6
-merge_method: dare_ties
-base_model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO
-parameters:
-int8_mask: true
-dtype: bfloat16
-random_seed: 0
-    ```
+This model was merged using the SLERP merge method.
 
-## 💻 Usage
+### Models Merged
 
-```python
-!pip install -qU transformers accelerate
+The following models were included in the merge:
+* [eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2](https://huggingface.co/eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2)
+* [yam-peleg/Experiment27-7B](https://huggingface.co/yam-peleg/Experiment27-7B)
 
-from transformers import AutoTokenizer
-import transformers
-import torch
+### Configuration
 
-model = "automerger/OgnoExperiment27-7B"
-messages = [{"role": "user", "content": "What is a large language model?"}]
+The following YAML configuration was used to produce this 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",
-)
+```yaml
 
-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"])
-```
+slices:
+  - sources:
+      - model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2
+        layer_range: [0, 32]
+      - model: yam-peleg/Experiment27-7B
+        layer_range: [0, 32]
+merge_method: slerp
+base_model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2
+parameters:
+  t:
+    - filter: self_attn
+      value: [0, 0.5, 0.3, 0.7, 1]
+    - filter: mlp
+      value: [1, 0.5, 0.7, 0.3, 0]
+    - value: 0.5
+dtype: bfloat16
+random_seed: 0
+    
+```

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO",
+  "_name_or_path": "eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2",
   "architectures": [
     "MistralForCausalLM"
   ],

mergekit_config.yml

@@ -1,15 +1,19 @@
 
-models:
-- model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO
-  # No parameters necessary for base model
-- model: yam-peleg/Experiment27-7B
-  parameters:
-    density: 0.53
-    weight: 0.6
-merge_method: dare_ties
-base_model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO
+slices:
+  - sources:
+      - model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2
+        layer_range: [0, 32]
+      - model: yam-peleg/Experiment27-7B
+        layer_range: [0, 32]
+merge_method: slerp
+base_model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2
 parameters:
-int8_mask: true
+  t:
+    - filter: self_attn
+      value: [0, 0.5, 0.3, 0.7, 1]
+    - filter: mlp
+      value: [1, 0.5, 0.7, 0.3, 0]
+    - value: 0.5
 dtype: bfloat16
 random_seed: 0
     

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4918cf334c476da5dd92d07afbfe6d9353a8bcbea0e724c8d97a9fadcebffe1a
+oid sha256:6da1ee81fbbe562364deabfe20418541b64e0abe9c5e8ab827bd3952bcc8a346
 size 9942981696

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:34373bd2a86f1c3a278da58ab2056680e831bb4aa1555b6ed52e4555716ba67b
+oid sha256:bb36def4e091c55991fade8802229da81250b57d49e892848d56a2f530a89f48
 size 4540516344