Upload folder using huggingface_hub

README.md

@@ -20,10 +20,10 @@ tags:
 #parameters:
 #  t:
 #    - filter: self_attn
-#      value: [0.32115546068984224, 0.8724514048921832, 0.9416438704146307, 0.13454654769720242, 0.0020171582354998607]
+#      value: [0.10671682864418519, 0.2587429682722914, 0.7254855120576987, 0.36811557880904033, 0.01600047241283964]
 #    - filter: mlp
-#      value: [0.6788445393101578, 0.1275485951078168, 0.8654534523027976, 0.8654534523027976, 0.9979828417645001]
-#    - value: 0.23694210240180313
+#      value: [0.8932831713558148, 0.7412570317277086, 0.6318844211909597, 0.6318844211909597, 0.9839995275871604]
+#    - value: 0.6580019418208543
 #dtype: bfloat16
 #random_seed: 0
 
@@ -38,10 +38,10 @@ tags:
 #parameters:
 #  t:
 #    - filter: self_attn
-#      value: [0.32115546068984224, 0.8724514048921832, 0.9416438704146307, 0.13454654769720242, 0.0020171582354998607]
+#      value: [0.10671682864418519, 0.2587429682722914, 0.7254855120576987, 0.36811557880904033, 0.01600047241283964]
 #    - filter: mlp
-#      value: [0.6788445393101578, 0.1275485951078168, 0.05835612958536929, 0.8654534523027976, 0.9979828417645001]
-#    - value: 0.23694210240180313
+#      value: [0.8932831713558148, 0.7412570317277086, 0.27451448794230127, 0.6318844211909597, 0.9839995275871604]
+#    - value: 0.6580019418208543
 #dtype: float16
 #random_seed: 0
 
@@ -79,7 +79,7 @@ experts:
     - "tell me"
     - "explain"
     - "I want"
-  - source_model: mlabonne/NeuralDaredevil-8B-abliterated
+  - source_model: TheBloke/Llama-2-13B-fp16
     positive_prompts:
     - "reason"
     - "math"

mergekit_moe_config.yml

@@ -10,10 +10,10 @@
 #parameters:
 #  t:
 #    - filter: self_attn
-#      value: [0.32115546068984224, 0.8724514048921832, 0.9416438704146307, 0.13454654769720242, 0.0020171582354998607]
+#      value: [0.10671682864418519, 0.2587429682722914, 0.7254855120576987, 0.36811557880904033, 0.01600047241283964]
 #    - filter: mlp
-#      value: [0.6788445393101578, 0.1275485951078168, 0.8654534523027976, 0.8654534523027976, 0.9979828417645001]
-#    - value: 0.23694210240180313
+#      value: [0.8932831713558148, 0.7412570317277086, 0.6318844211909597, 0.6318844211909597, 0.9839995275871604]
+#    - value: 0.6580019418208543
 #dtype: bfloat16
 #random_seed: 0
 
@@ -28,10 +28,10 @@
 #parameters:
 #  t:
 #    - filter: self_attn
-#      value: [0.32115546068984224, 0.8724514048921832, 0.9416438704146307, 0.13454654769720242, 0.0020171582354998607]
+#      value: [0.10671682864418519, 0.2587429682722914, 0.7254855120576987, 0.36811557880904033, 0.01600047241283964]
 #    - filter: mlp
-#      value: [0.6788445393101578, 0.1275485951078168, 0.05835612958536929, 0.8654534523027976, 0.9979828417645001]
-#    - value: 0.23694210240180313
+#      value: [0.8932831713558148, 0.7412570317277086, 0.27451448794230127, 0.6318844211909597, 0.9839995275871604]
+#    - value: 0.6580019418208543
 #dtype: float16
 #random_seed: 0
 
@@ -69,7 +69,7 @@ experts:
     - "tell me"
     - "explain"
     - "I want"
-  - source_model: mlabonne/NeuralDaredevil-8B-abliterated
+  - source_model: TheBloke/Llama-2-13B-fp16
     positive_prompts:
     - "reason"
     - "math"

model-00001-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cba494aca2a9f9a1ec5f9a1e250f7dbc8ce981b53f279654d8c9e8df7d9c45a7
+size 1936995488

model-00002-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86514b4aa7092ded8aec85def4de7edde4ffc1b875cbb6e5f61e3026403427aa
+size 1954547640

model-00003-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37f244fa57b98f26b2c3a36c45929d67e636fa76b48b4df377c2a49ebbb6ff2b
+size 1930430400

model-00004-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ec4b742d3d134f375ce5dc5ad6652f3eff7cd02ee873e097e2640b86a10cde6
+size 1954547648

model-00005-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:531640ca9121569bae7d6cf670cffc966ef353b4dbe9e3ddeae5bafd3c8bf842
+size 1930430392

model-00006-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0f1046734d6e492519192aba885e63ff9e3127e954cae69186f297de32134e7
+size 1954547640

model-00007-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8e715e48a666161b0b87b92c377fef989da135049440aa9b43ce17fa82b53ce
+size 1930430400

model-00008-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1be764073653fe62d8b0a99de6d9ffda166f4ba61f0e04ee1f5b6d12f00b0e99
+size 1954547648

model-00009-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f79041dc8347cd6473276d621601a333859ec985bda3820b537ebec970769ef
+size 1930430400

model-00010-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c66bb7d3cdbbe358b28a66e4e8bbfec9124d6f4588e07ec66f6afe0d39505df4
+size 1954547632

model-00011-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11229579db1f5f9b326277f4398b4bd1547347630189075d21098f577e27a709
+size 1930430400

model-00012-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:322044e13ea95ffd8b3391556fc47a40e7b045f1953878eda61e08eff174c531
+size 1954547648

model-00013-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3894ef4784b4e37eec260fe90f26045478bfb1bbd89eb031f6c09015a82e8945
+size 1930430400

model-00014-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35aeeb70e9dc703b47aef1e7393175a89982f4dbe07081ec4121e5c4ee03d275
+size 1987329504

model-00015-of-00015.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f906c86e8f8781d4d8adb64fdb827d709fe8b5ee439095640791e14b53bdce4
+size 839391696

model.safetensors.index.json

@@ -1 +1 @@
-{"metadata": {"mergekit_version": "0.0.4"}, "weight_map": {"model.embed_tokens.weight": "model-00001-of-00013.safetensors", "model.norm.weight": "model-00001-of-00013.safetensors", "lm_head.weight": "model-00001-of-00013.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.5.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.6.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.7.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.8.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.9.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.11.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.12.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.13.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.14.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.15.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.16.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.17.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.18.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.19.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.20.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.21.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.22.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.23.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.24.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.25.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.26.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.27.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.28.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.29.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.30.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.31.input_layernorm.weight": "model-00001-of-00013.safetensors", "model.layers.0.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00013.safetensors", "model.layers.0.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00013.safetensors", "model.layers.1.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00013.safetensors", "model.layers.1.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00013.safetensors", "model.layers.2.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00013.safetensors", "model.layers.2.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00013.safetensors", "model.layers.3.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00013.safetensors", "model.layers.3.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00013.safetensors", "model.layers.4.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00013.safetensors", "model.layers.4.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00013.safetensors", "model.layers.5.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00013.safetensors", "model.layers.5.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00013.safetensors", "model.layers.6.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.6.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.7.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.7.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.8.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.8.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.9.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.9.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.10.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.10.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.11.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.11.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.12.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.12.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.13.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.13.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00013.safetensors", "model.layers.14.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00013.safetensors", "model.layers.14.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.15.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.15.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.16.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.16.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.17.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.17.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.18.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.18.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.19.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.19.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.20.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.20.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.21.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.21.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.22.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00013.safetensors", "model.layers.22.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00013.safetensors", "model.layers.23.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.23.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.24.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.24.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.25.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.25.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.26.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.26.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.27.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.27.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.28.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.28.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.29.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.29.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.30.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.30.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00013.safetensors", "model.layers.31.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00013.safetensors", "model.layers.31.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00013.safetensors", "model.layers.0.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.0.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.1.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.1.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.2.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.2.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.3.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.3.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.4.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.4.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.5.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.5.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.6.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.6.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.7.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00013.safetensors", "model.layers.7.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00013.safetensors", "model.layers.8.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.8.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.9.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.9.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.10.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.10.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.11.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.11.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.12.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.12.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.13.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.13.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.14.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.14.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.15.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.15.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00013.safetensors", "model.layers.16.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00013.safetensors", "model.layers.16.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.17.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.17.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.18.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.18.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.19.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.19.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.20.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.20.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.21.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.21.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.22.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.22.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.23.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.23.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.24.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00013.safetensors", "model.layers.24.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00013.safetensors", "model.layers.25.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00013.safetensors", "model.layers.25.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00013.safetensors", "model.layers.26.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00013.safetensors", "model.layers.26.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00013.safetensors", "model.layers.27.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00013.safetensors", "model.layers.27.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00013.safetensors", "model.layers.28.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00013.safetensors", "model.layers.28.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00013.safetensors", "model.layers.29.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00013.safetensors", "model.layers.29.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00013.safetensors", "model.layers.30.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00013.safetensors", "model.layers.30.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00013.safetensors", "model.layers.31.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00013.safetensors", "model.layers.31.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00013.safetensors", "model.layers.0.block_sparse_moe.experts.0.w1.weight": "model-00008-of-00013.safetensors", "model.layers.0.block_sparse_moe.experts.1.w1.weight": "model-00008-of-00013.safetensors", "model.layers.1.block_sparse_moe.experts.0.w1.weight": "model-00008-of-00013.safetensors", "model.layers.1.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.2.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.2.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.3.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.3.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.4.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.4.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.5.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.5.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.6.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.6.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.7.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.7.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.8.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.8.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.9.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00013.safetensors", "model.layers.9.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00013.safetensors", "model.layers.10.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.10.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.11.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.11.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.12.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.12.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.13.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.13.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.14.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.14.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.15.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.15.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.16.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.16.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.17.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.17.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00013.safetensors", "model.layers.18.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00013.safetensors", "model.layers.18.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.19.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.19.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.20.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.20.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.21.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.21.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.22.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.22.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.23.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.23.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.24.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.24.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.25.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.25.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.26.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00013.safetensors", "model.layers.26.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00013.safetensors", "model.layers.27.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00013.safetensors", "model.layers.27.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00013.safetensors", "model.layers.28.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00013.safetensors", "model.layers.28.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00013.safetensors", "model.layers.29.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00013.safetensors", "model.layers.29.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00013.safetensors", "model.layers.30.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00013.safetensors", "model.layers.30.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00013.safetensors", "model.layers.31.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00013.safetensors", "model.layers.31.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00013.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00012-of-00013.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00012-of-00013.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00013-of-00013.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00013-of-00013.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00013-of-00013.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00013-of-00013.safetensors", "model.layers.0.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.1.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.2.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.3.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.4.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.5.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.6.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.7.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.8.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.9.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.10.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.11.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.12.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.13.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.14.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.15.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.16.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.17.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.18.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.19.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.20.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.21.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.22.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.23.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.24.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.25.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.26.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.27.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.28.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.29.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.30.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors", "model.layers.31.block_sparse_moe.gate.weight": "model-00013-of-00013.safetensors"}}
+{"metadata": {"mergekit_version": "0.0.4"}, "weight_map": {"model.embed_tokens.weight": "model-00001-of-00015.safetensors", "model.norm.weight": "model-00001-of-00015.safetensors", "lm_head.weight": "model-00001-of-00015.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.5.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.6.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.7.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.8.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.9.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.11.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.12.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.13.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.14.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.15.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.16.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.17.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.18.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.19.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.20.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.21.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.22.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.23.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.24.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.25.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.26.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.27.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.28.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.29.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.30.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.31.input_layernorm.weight": "model-00001-of-00015.safetensors", "model.layers.0.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00015.safetensors", "model.layers.0.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00015.safetensors", "model.layers.1.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00015.safetensors", "model.layers.1.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00015.safetensors", "model.layers.2.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00015.safetensors", "model.layers.2.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00015.safetensors", "model.layers.3.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00015.safetensors", "model.layers.3.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00015.safetensors", "model.layers.4.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00015.safetensors", "model.layers.4.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00015.safetensors", "model.layers.5.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00015.safetensors", "model.layers.5.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.6.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00015.safetensors", "model.layers.6.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.7.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00015.safetensors", "model.layers.7.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.8.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00015.safetensors", "model.layers.8.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.9.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00015.safetensors", "model.layers.9.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.10.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00015.safetensors", "model.layers.10.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.11.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00015.safetensors", "model.layers.11.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.12.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00015.safetensors", "model.layers.12.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00015.safetensors", "model.layers.13.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.13.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00015.safetensors", "model.layers.14.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.14.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00015.safetensors", "model.layers.15.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.15.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00015.safetensors", "model.layers.16.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.16.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00015.safetensors", "model.layers.17.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.17.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00015.safetensors", "model.layers.18.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.18.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00015.safetensors", "model.layers.19.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.19.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00015.safetensors", "model.layers.20.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00015.safetensors", "model.layers.20.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.21.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00015.safetensors", "model.layers.21.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.22.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00015.safetensors", "model.layers.22.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.23.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00015.safetensors", "model.layers.23.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.24.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00015.safetensors", "model.layers.24.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.25.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00015.safetensors", "model.layers.25.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.26.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00015.safetensors", "model.layers.26.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.27.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00015.safetensors", "model.layers.27.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00015.safetensors", "model.layers.28.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00015.safetensors", "model.layers.28.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00015.safetensors", "model.layers.29.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00015.safetensors", "model.layers.29.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00015.safetensors", "model.layers.30.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00015.safetensors", "model.layers.30.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00015.safetensors", "model.layers.31.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00015.safetensors", "model.layers.31.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00015.safetensors", "model.layers.0.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00015.safetensors", "model.layers.0.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00015.safetensors", "model.layers.1.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00015.safetensors", "model.layers.1.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00015.safetensors", "model.layers.2.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00015.safetensors", "model.layers.2.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00015.safetensors", "model.layers.3.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00015.safetensors", "model.layers.3.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.4.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00015.safetensors", "model.layers.4.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.5.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00015.safetensors", "model.layers.5.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.6.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00015.safetensors", "model.layers.6.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.7.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00015.safetensors", "model.layers.7.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.8.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00015.safetensors", "model.layers.8.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.9.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00015.safetensors", "model.layers.9.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.10.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00015.safetensors", "model.layers.10.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00015.safetensors", "model.layers.11.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.11.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00015.safetensors", "model.layers.12.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.12.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00015.safetensors", "model.layers.13.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.13.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00015.safetensors", "model.layers.14.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.14.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00015.safetensors", "model.layers.15.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.15.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00015.safetensors", "model.layers.16.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.16.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00015.safetensors", "model.layers.17.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.17.block_sparse_moe.experts.1.w2.weight": "model-00007-of-00015.safetensors", "model.layers.18.block_sparse_moe.experts.0.w2.weight": "model-00007-of-00015.safetensors", "model.layers.18.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.19.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00015.safetensors", "model.layers.19.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.20.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00015.safetensors", "model.layers.20.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.21.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00015.safetensors", "model.layers.21.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.22.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00015.safetensors", "model.layers.22.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.23.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00015.safetensors", "model.layers.23.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.24.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00015.safetensors", "model.layers.24.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.25.block_sparse_moe.experts.0.w2.weight": "model-00008-of-00015.safetensors", "model.layers.25.block_sparse_moe.experts.1.w2.weight": "model-00008-of-00015.safetensors", "model.layers.26.block_sparse_moe.experts.0.w2.weight": "model-00009-of-00015.safetensors", "model.layers.26.block_sparse_moe.experts.1.w2.weight": "model-00009-of-00015.safetensors", "model.layers.27.block_sparse_moe.experts.0.w2.weight": "model-00009-of-00015.safetensors", "model.layers.27.block_sparse_moe.experts.1.w2.weight": "model-00009-of-00015.safetensors", "model.layers.28.block_sparse_moe.experts.0.w2.weight": "model-00009-of-00015.safetensors", "model.layers.28.block_sparse_moe.experts.1.w2.weight": "model-00009-of-00015.safetensors", "model.layers.29.block_sparse_moe.experts.0.w2.weight": "model-00009-of-00015.safetensors", "model.layers.29.block_sparse_moe.experts.1.w2.weight": "model-00009-of-00015.safetensors", "model.layers.30.block_sparse_moe.experts.0.w2.weight": "model-00009-of-00015.safetensors", "model.layers.30.block_sparse_moe.experts.1.w2.weight": "model-00009-of-00015.safetensors", "model.layers.31.block_sparse_moe.experts.0.w2.weight": "model-00009-of-00015.safetensors", "model.layers.31.block_sparse_moe.experts.1.w2.weight": "model-00009-of-00015.safetensors", "model.layers.0.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00015.safetensors", "model.layers.0.block_sparse_moe.experts.1.w1.weight": "model-00009-of-00015.safetensors", "model.layers.1.block_sparse_moe.experts.0.w1.weight": "model-00009-of-00015.safetensors", "model.layers.1.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.2.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00015.safetensors", "model.layers.2.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.3.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00015.safetensors", "model.layers.3.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.4.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00015.safetensors", "model.layers.4.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.5.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00015.safetensors", "model.layers.5.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.6.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00015.safetensors", "model.layers.6.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.7.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00015.safetensors", "model.layers.7.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.8.block_sparse_moe.experts.0.w1.weight": "model-00010-of-00015.safetensors", "model.layers.8.block_sparse_moe.experts.1.w1.weight": "model-00010-of-00015.safetensors", "model.layers.9.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.9.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00015.safetensors", "model.layers.10.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.10.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00015.safetensors", "model.layers.11.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.11.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00015.safetensors", "model.layers.12.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.12.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00015.safetensors", "model.layers.13.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.13.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00015.safetensors", "model.layers.14.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.14.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00015.safetensors", "model.layers.15.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.15.block_sparse_moe.experts.1.w1.weight": "model-00011-of-00015.safetensors", "model.layers.16.block_sparse_moe.experts.0.w1.weight": "model-00011-of-00015.safetensors", "model.layers.16.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.17.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00015.safetensors", "model.layers.17.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.18.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00015.safetensors", "model.layers.18.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.19.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00015.safetensors", "model.layers.19.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.20.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00015.safetensors", "model.layers.20.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.21.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00015.safetensors", "model.layers.21.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.22.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00015.safetensors", "model.layers.22.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.23.block_sparse_moe.experts.0.w1.weight": "model-00012-of-00015.safetensors", "model.layers.23.block_sparse_moe.experts.1.w1.weight": "model-00012-of-00015.safetensors", "model.layers.24.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.24.block_sparse_moe.experts.1.w1.weight": "model-00013-of-00015.safetensors", "model.layers.25.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.25.block_sparse_moe.experts.1.w1.weight": "model-00013-of-00015.safetensors", "model.layers.26.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.26.block_sparse_moe.experts.1.w1.weight": "model-00013-of-00015.safetensors", "model.layers.27.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.27.block_sparse_moe.experts.1.w1.weight": "model-00013-of-00015.safetensors", "model.layers.28.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.28.block_sparse_moe.experts.1.w1.weight": "model-00013-of-00015.safetensors", "model.layers.29.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.29.block_sparse_moe.experts.1.w1.weight": "model-00013-of-00015.safetensors", "model.layers.30.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.30.block_sparse_moe.experts.1.w1.weight": "model-00013-of-00015.safetensors", "model.layers.31.block_sparse_moe.experts.0.w1.weight": "model-00013-of-00015.safetensors", "model.layers.31.block_sparse_moe.experts.1.w1.weight": "model-00014-of-00015.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00014-of-00015.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00014-of-00015.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00014-of-00015.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00014-of-00015.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00014-of-00015.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00014-of-00015.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00014-of-00015.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00014-of-00015.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00014-of-00015.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00014-of-00015.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00014-of-00015.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00015-of-00015.safetensors", "model.layers.0.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.1.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.2.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.3.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.4.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.5.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.6.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.7.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.8.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.9.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.10.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.11.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.12.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.13.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.14.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.15.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.16.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.17.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.18.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.19.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.20.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.21.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.22.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.23.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.24.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.25.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.26.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.27.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.28.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.29.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.30.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors", "model.layers.31.block_sparse_moe.gate.weight": "model-00015-of-00015.safetensors"}}