Upload folder using huggingface_hub

README.md

@@ -0,0 +1,48 @@
+---
+base_model:
+- Qwen/Qwen3-32B
+- LLMcompe-Team-Watanabe/Qwen3-32B-sft-Deepmath-8k
+- LLMcompe-Team-Watanabe/Qwen3-32B-openmathreasoning-sft
+library_name: transformers
+tags:
+- mergekit
+- merge
+
+---
+# output_model
+
+This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
+
+## Merge Details
+### Merge Method
+
+This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using [Qwen/Qwen3-32B](https://huggingface.co/Qwen/Qwen3-32B) as a base.
+
+### Models Merged
+
+The following models were included in the merge:
+* [LLMcompe-Team-Watanabe/Qwen3-32B-sft-Deepmath-8k](https://huggingface.co/LLMcompe-Team-Watanabe/Qwen3-32B-sft-Deepmath-8k)
+* [LLMcompe-Team-Watanabe/Qwen3-32B-openmathreasoning-sft](https://huggingface.co/LLMcompe-Team-Watanabe/Qwen3-32B-openmathreasoning-sft)
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+models:
+  - model: Qwen/Qwen3-32B
+  - model: LLMcompe-Team-Watanabe/Qwen3-32B-sft-Deepmath-8k
+    parameters:
+      density: 0.53
+      weight: 0.3
+  - model: LLMcompe-Team-Watanabe/Qwen3-32B-openmathreasoning-sft
+    parameters:
+      density: 0.53
+      weight: 0.7
+merge_method: dare_ties
+base_model: Qwen/Qwen3-32B
+parameters:
+  int8_mask: true
+  normalize: false
+dtype: bfloat16 
+```

config.json

@@ -11,6 +11,72 @@
   "hidden_size": 5120,
   "initializer_range": 0.02,
   "intermediate_size": 25600,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
   "max_position_embeddings": 40960,
   "max_window_layers": 64,
   "model_type": "qwen3",
@@ -22,9 +88,9 @@
   "rope_theta": 1000000,
   "sliding_window": null,
   "tie_word_embeddings": false,
-  "torch_dtype": "float32",
-  "transformers_version": "4.52.3",
-  "use_cache": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.1",
+  "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 151936
 }

mergekit_config.yml

@@ -0,0 +1,16 @@
+models:
+  - model: Qwen/Qwen3-32B
+  - model: LLMcompe-Team-Watanabe/Qwen3-32B-sft-Deepmath-8k
+    parameters:
+      density: 0.53
+      weight: 0.3
+  - model: LLMcompe-Team-Watanabe/Qwen3-32B-openmathreasoning-sft
+    parameters:
+      density: 0.53
+      weight: 0.7
+merge_method: dare_ties
+base_model: Qwen/Qwen3-32B
+parameters:
+  int8_mask: true
+  normalize: false
+dtype: bfloat16 

model-00001-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8c8bb6cb9dc1003c6e4af3f6f0d900c545c2f6dd030e033038af6a6398e3b08
-size 9864610392
+oid sha256:f02e5c45e8a1da8ea5222447c2e2394f430c26f8106c3ff86dd16090fb6e0f92
+size 4967673312

model-00002-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:469bbac1f8e251e0ba81c7a2b4d40f3e2edddcfac7010a4c39317496a99fe337
-size 9751973008
+oid sha256:7a56c6a1a494b33442e0eb60189b66d5f89d038552becd8b350c293c7c27827d
+size 4970372168

model-00003-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f32327088962a3b05d74f48a812c8a74590cd14302c79aed761e56d235db2fcd
-size 9751973048
+oid sha256:e559eaef8dc0e6a990140599cd05ec5fc8ce66f54328196c16e33c44f07f5b57
+size 4875989744

model-00004-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f84268ab38347ba32e471cde8dafeb148993705f6506123169cf56b5c0dba1cb
-size 9751973064
+oid sha256:8606e6def683c91cfdb8985549402d5accdc0ec95b6238ab1a6d565aa775d8b3
+size 4875989736

model-00005-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0db21de71dc685d6499b191ee241a2dca378e3180d50024ffbe35b7b73b782db
-size 9751973064
+oid sha256:988edf3943a736a43efad8364ff4932b1e57c2622f5378d04a5073ad75b81b61
+size 4875989744

model-00006-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d7763611e952ab17d31975a989a40c3135ce102daf7b4a35bc85db4729ad1fb
-size 9751973064
+oid sha256:898d8f51ac42ba81f2b663112cefd08181882d931d73718f9f8366cfaef0521c
+size 4875989736

model-00007-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84839c17e10a845423e1b822c0da3523555a67b3c281c6de802b463aa76f5d4f
-size 9751973064
+oid sha256:befba4dd98109f16acb002f1ee108cc9daf9cc6fc693d739bf7d6962af563966
+size 4875989744

model-00008-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc09d634bd97e0e77b2dd90733677f11b487dc69774ea3ff25986084c9599c1d
-size 9751973064
+oid sha256:279d74546e4c37b5c963dac97435bb1ea7875a59a205b9fa7a8125cdee2f36ae
+size 4875989736

model-00009-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c7666095337d23897ee42e3c2e4e3c72f1d98eaf05c689d4701f372d3c3308e4
-size 9751973064
+oid sha256:63ce0b032fdb81b248f0849be258ed68ec724414085c6c13c9d6b833d8afea6c
+size 4875989744

model-00010-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8dea39d3cc75a4934a3a3f4813c07212fe365bdde37fdb9e679a637aa0b9577c
-size 9751973064
+oid sha256:e27c8e246391cf432e037b14e6cbd0ed995df85be14168004224c6ed30dabc41
+size 4875989736

model-00011-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd63e860745c27257230ca0d2897b0169df64dfb1400c947b1772efd1bd0539d
-size 9751973064
+oid sha256:d87730f03f429d31f49eefd9b11e4fb8d0d59c272c328ab3d1f743b8332c88bf
+size 4875989744

model-00012-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9283f11468335bd8754155b49594e67523464587127f4c4022b9f3b45af55cc
-size 9751973064
+oid sha256:9289e60f8231f2ff39d00e8b358600f9baa48d5952050b78bd0e092641e9aec8
+size 4875989736

model-00013-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba2ae9436ee9fb867aed29921adeda4865363cd2de01ce9d01ac97b9a8c07c6c
-size 9751973064
+oid sha256:28d9453738713a50fa02c3de8839290357fe8ba84934e6c881e2e9e04500d619
+size 4875989736

model-00014-of-00014.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa058b53557684e1f131c0b285376c4c2a2c8149d7fbda7cbc06e40b758b4c5e
-size 4160287400
+oid sha256:762cd241341853c91b2e3f8fe97acef067e79c00c1ac8db8415e68fa9a30e5fc
+size 1950395864

tokenizer_config.json

@@ -227,10 +227,10 @@
     "<|video_pad|>"
   ],
   "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if message.content is string %}\n        {%- set content = message.content %}\n    {%- else %}\n        {%- set content = '' %}\n    {%- endif %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is string %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in content %}\n                {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n                {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",
-  "extra_special_tokens": {},
   "model_max_length": 131072,
   "pad_token": "<|endoftext|>",
   "split_special_tokens": false,