Upload folder using huggingface_hub

README.md

@@ -3,38 +3,52 @@ tags:
 - merge
 - mergekit
 - lazymergekit
-- upaya07/Arithmo2-Mistral-7B
 - WizardLMTeam/WizardMath-7B-V1.1
+- microsoft/rho-math-7b-interpreter-v0.1
+- meta-math/MetaMath-Mistral-7B
 base_model:
-- upaya07/Arithmo2-Mistral-7B
 - WizardLMTeam/WizardMath-7B-V1.1
+- microsoft/rho-math-7b-interpreter-v0.1
+- meta-math/MetaMath-Mistral-7B
 ---
 
 # ganeet-V4
 
 ganeet-V4 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
-* [upaya07/Arithmo2-Mistral-7B](https://huggingface.co/upaya07/Arithmo2-Mistral-7B)
 * [WizardLMTeam/WizardMath-7B-V1.1](https://huggingface.co/WizardLMTeam/WizardMath-7B-V1.1)
+* [microsoft/rho-math-7b-interpreter-v0.1](https://huggingface.co/microsoft/rho-math-7b-interpreter-v0.1)
+* [meta-math/MetaMath-Mistral-7B](https://huggingface.co/meta-math/MetaMath-Mistral-7B)
 
 ## 🧩 Configuration
 
 ```yaml
-slices:
-  - sources:
-      - model: upaya07/Arithmo2-Mistral-7B
-        layer_range: [0, 32]
-      - model: WizardLMTeam/WizardMath-7B-V1.1
-        layer_range: [0, 32]
-merge_method: slerp
-base_model: upaya07/Arithmo2-Mistral-7B
+models:
+  - model: WizardLMTeam/WizardMath-7B-V1.1 
+    layer_range: [0, 30]
+    parameters:
+      density: 0.5  # fraction of weights in differences from the base model to retain
+      weight:   # weight gradient
+        - filter: mlp
+          value: 0.5
+        - value: 0
+  - model: deepseek-ai/deepseek-math-7b-rl
+    layer_range: [0, 30]
+  - model: microsoft/rho-math-7b-interpreter-v0.1
+    layer_range: [0, 30]
+    parameters:
+      density: 0.5
+      weight: 0.5
+  - model: meta-math/MetaMath-Mistral-7B
+    layer_range: [0, 30]
+    parameters:
+      density: 0.5
+      weight: 0.5
+merge_method: ties
+base_model: deepseek-ai/deepseek-math-7b-rl
 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: 0.4
-dtype: bfloat16
+  normalize: true
+  int8_mask: true
+dtype: float16
 ```
 
 ## 💻 Usage