Upload folder using huggingface_hub

README.md

@@ -1,5 +1,5 @@
 ---
-license: apache-2.0
+license: cc-by-nc-4.0
 tags:
 - merge
 - mergekit
@@ -7,36 +7,30 @@ tags:
 - automerger
 base_model:
 - rwitz/experiment26-truthy-iter-0
-- yam-peleg/Experiment26-7B
 ---
 
 # Experiment26Experiment26-7B
 
 Experiment26Experiment26-7B is an automated merge created by [Maxime Labonne](https://huggingface.co/mlabonne) using the following configuration.
 * [rwitz/experiment26-truthy-iter-0](https://huggingface.co/rwitz/experiment26-truthy-iter-0)
-* [yam-peleg/Experiment26-7B](https://huggingface.co/yam-peleg/Experiment26-7B)
 
 ## 🧩 Configuration
 
 ```yaml
-slices:
-  - sources:
-      - model: rwitz/experiment26-truthy-iter-0
-        layer_range: [0, 32]
-      - model: yam-peleg/Experiment26-7B
-        layer_range: [0, 32]
-merge_method: slerp
-base_model: rwitz/experiment26-truthy-iter-0
+models:
+- model: rwitz/experiment26-truthy-iter-2
+  # No parameters necessary for base model
+- model: rwitz/experiment26-truthy-iter-0
+  parameters:
+    density: 0.53
+    weight: 0.6
+merge_method: dare_ties
+base_model: rwitz/experiment26-truthy-iter-2
 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
+int8_mask: true
 dtype: bfloat16
 random_seed: 0
-    ```
+```
 
 ## 💻 Usage
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "rwitz/experiment26-truthy-iter-0",
+  "_name_or_path": "rwitz/experiment26-truthy-iter-2",
   "architectures": [
     "MistralForCausalLM"
   ],
@@ -20,7 +20,7 @@
   "sliding_window": 4096,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.39.0",
+  "transformers_version": "4.39.1",
   "use_cache": true,
   "vocab_size": 32000
 }

mergekit_config.yml

@@ -1,19 +1,14 @@
 
-slices:
-  - sources:
-      - model: rwitz/experiment26-truthy-iter-0
-        layer_range: [0, 32]
-      - model: yam-peleg/Experiment26-7B
-        layer_range: [0, 32]
-merge_method: slerp
-base_model: rwitz/experiment26-truthy-iter-0
+models:
+- model: rwitz/experiment26-truthy-iter-2
+  # No parameters necessary for base model
+- model: rwitz/experiment26-truthy-iter-0
+  parameters:
+    density: 0.53
+    weight: 0.6
+merge_method: dare_ties
+base_model: rwitz/experiment26-truthy-iter-2
 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
+int8_mask: true
 dtype: bfloat16
 random_seed: 0
-    

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c2040230cbe78f2d058b7866da799b07a846eb7bb8ae405b4a29330bf44e587
+oid sha256:3431f9f8db214deb3f8b9c39e69bb5ce2081850f7c110941cbab3143f150e370
 size 9942981696

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:135770e8e6aed94e95184c999e8e081160cf9d0f551353d8565cc510ede4e382
+oid sha256:cc32ab43dea4a51d8c2f9d5aa4ad4ea20199bab5c5041c52f645b90d7b556347
 size 4540516344