Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,54 @@
+---
+license: apache-2.0
+base_model: DavidAU/Qwen2.5-2X7B-Coder-CodeV-R1-Coder-Instruct-OlympicCoder-19B
+language:
+- en
+pipeline_tag: text-generation
+tags:
+- merge
+- programming
+- code generation
+- code
+- codeqwen
+- moe
+- coding
+- coder
+- qwen2
+- chat
+- qwen
+- qwen-coder
+- mixture of experts
+- qwen2moe
+- 2X7B
+- shared expert
+- mlx
+library_name: mlx
+---
+
+# Qwen2.5-2X7B-Coder-CodeV-R1-Coder-Instruct-OlympicCoder-19B-dq68-mlx
+
+This model [Qwen2.5-2X7B-Coder-CodeV-R1-Coder-Instruct-OlympicCoder-19B-dq68-mlx](https://huggingface.co/Qwen2.5-2X7B-Coder-CodeV-R1-Coder-Instruct-OlympicCoder-19B-dq68-mlx) was
+converted to MLX format from [DavidAU/Qwen2.5-2X7B-Coder-CodeV-R1-Coder-Instruct-OlympicCoder-19B](https://huggingface.co/DavidAU/Qwen2.5-2X7B-Coder-CodeV-R1-Coder-Instruct-OlympicCoder-19B)
+using mlx-lm version **0.26.0**.
+
+## Use with mlx
+
+```bash
+pip install mlx-lm
+```
+
+```python
+from mlx_lm import load, generate
+
+model, tokenizer = load("Qwen2.5-2X7B-Coder-CodeV-R1-Coder-Instruct-OlympicCoder-19B-dq68-mlx")
+
+prompt = "hello"
+
+if tokenizer.chat_template is not None:
+    messages = [{"role": "user", "content": prompt}]
+    prompt = tokenizer.apply_chat_template(
+        messages, add_generation_prompt=True
+    )
+
+response = generate(model, tokenizer, prompt=prompt, verbose=True)
+```

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\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>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\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" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

config.json

@@ -0,0 +1,956 @@
+{
+    "architectures": [
+        "Qwen2MoeForCausalLM"
+    ],
+    "attention_dropout": 0.0,
+    "decoder_sparse_step": 1,
+    "embd_pdrop": 0.0,
+    "eos_token_id": 151645,
+    "hidden_act": "silu",
+    "hidden_size": 3584,
+    "initializer_range": 0.02,
+    "intermediate_size": 18944,
+    "max_position_embeddings": 32768,
+    "max_window_layers": 28,
+    "mlp_only_layers": [],
+    "model_type": "qwen2_moe",
+    "moe_intermediate_size": 18944,
+    "norm_topk_prob": true,
+    "num_attention_heads": 28,
+    "num_experts": 2,
+    "num_experts_per_tok": 2,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 4,
+    "output_router_logits": false,
+    "pad_token_id": 151643,
+    "qkv_bias": true,
+    "quantization": {
+        "group_size": 32,
+        "bits": 6,
+        "model.embed_tokens": true,
+        "model.layers.0.self_attn.q_proj": true,
+        "model.layers.0.self_attn.k_proj": true,
+        "model.layers.0.self_attn.v_proj": true,
+        "model.layers.0.self_attn.o_proj": true,
+        "model.layers.0.self_attn.rope": false,
+        "model.layers.0.mlp.gate": true,
+        "model.layers.0.mlp.switch_mlp.gate_proj": true,
+        "model.layers.0.mlp.switch_mlp.up_proj": true,
+        "model.layers.0.mlp.switch_mlp.down_proj": true,
+        "model.layers.0.mlp.switch_mlp.activation": false,
+        "model.layers.0.mlp.shared_expert.gate_proj": true,
+        "model.layers.0.mlp.shared_expert.down_proj": true,
+        "model.layers.0.mlp.shared_expert.up_proj": true,
+        "model.layers.0.mlp.shared_expert_gate": true,
+        "model.layers.0.input_layernorm": false,
+        "model.layers.0.post_attention_layernorm": false,
+        "model.layers.1.self_attn.q_proj": true,
+        "model.layers.1.self_attn.k_proj": true,
+        "model.layers.1.self_attn.v_proj": true,
+        "model.layers.1.self_attn.o_proj": true,
+        "model.layers.1.self_attn.rope": false,
+        "model.layers.1.mlp.gate": true,
+        "model.layers.1.mlp.switch_mlp.gate_proj": true,
+        "model.layers.1.mlp.switch_mlp.up_proj": true,
+        "model.layers.1.mlp.switch_mlp.down_proj": true,
+        "model.layers.1.mlp.switch_mlp.activation": false,
+        "model.layers.1.mlp.shared_expert.gate_proj": true,
+        "model.layers.1.mlp.shared_expert.down_proj": true,
+        "model.layers.1.mlp.shared_expert.up_proj": true,
+        "model.layers.1.mlp.shared_expert_gate": true,
+        "model.layers.1.input_layernorm": false,
+        "model.layers.1.post_attention_layernorm": false,
+        "model.layers.2.self_attn.q_proj": true,
+        "model.layers.2.self_attn.k_proj": true,
+        "model.layers.2.self_attn.v_proj": true,
+        "model.layers.2.self_attn.o_proj": true,
+        "model.layers.2.self_attn.rope": false,
+        "model.layers.2.mlp.gate": true,
+        "model.layers.2.mlp.switch_mlp.gate_proj": true,
+        "model.layers.2.mlp.switch_mlp.up_proj": true,
+        "model.layers.2.mlp.switch_mlp.down_proj": true,
+        "model.layers.2.mlp.switch_mlp.activation": false,
+        "model.layers.2.mlp.shared_expert.gate_proj": true,
+        "model.layers.2.mlp.shared_expert.down_proj": true,
+        "model.layers.2.mlp.shared_expert.up_proj": true,
+        "model.layers.2.mlp.shared_expert_gate": true,
+        "model.layers.2.input_layernorm": false,
+        "model.layers.2.post_attention_layernorm": false,
+        "model.layers.3.self_attn.q_proj": true,
+        "model.layers.3.self_attn.k_proj": true,
+        "model.layers.3.self_attn.v_proj": true,
+        "model.layers.3.self_attn.o_proj": true,
+        "model.layers.3.self_attn.rope": false,
+        "model.layers.3.mlp.gate": true,
+        "model.layers.3.mlp.switch_mlp.gate_proj": true,
+        "model.layers.3.mlp.switch_mlp.up_proj": true,
+        "model.layers.3.mlp.switch_mlp.down_proj": true,
+        "model.layers.3.mlp.switch_mlp.activation": false,
+        "model.layers.3.mlp.shared_expert.gate_proj": true,
+        "model.layers.3.mlp.shared_expert.down_proj": true,
+        "model.layers.3.mlp.shared_expert.up_proj": true,
+        "model.layers.3.mlp.shared_expert_gate": true,
+        "model.layers.3.input_layernorm": false,
+        "model.layers.3.post_attention_layernorm": false,
+        "model.layers.4.self_attn.q_proj": true,
+        "model.layers.4.self_attn.k_proj": true,
+        "model.layers.4.self_attn.v_proj": true,
+        "model.layers.4.self_attn.o_proj": true,
+        "model.layers.4.self_attn.rope": false,
+        "model.layers.4.mlp.gate": true,
+        "model.layers.4.mlp.switch_mlp.gate_proj": true,
+        "model.layers.4.mlp.switch_mlp.up_proj": true,
+        "model.layers.4.mlp.switch_mlp.down_proj": true,
+        "model.layers.4.mlp.switch_mlp.activation": false,
+        "model.layers.4.mlp.shared_expert.gate_proj": true,
+        "model.layers.4.mlp.shared_expert.down_proj": true,
+        "model.layers.4.mlp.shared_expert.up_proj": true,
+        "model.layers.4.mlp.shared_expert_gate": true,
+        "model.layers.4.input_layernorm": false,
+        "model.layers.4.post_attention_layernorm": false,
+        "model.layers.5.self_attn.q_proj": true,
+        "model.layers.5.self_attn.k_proj": true,
+        "model.layers.5.self_attn.v_proj": true,
+        "model.layers.5.self_attn.o_proj": true,
+        "model.layers.5.self_attn.rope": false,
+        "model.layers.5.mlp.gate": true,
+        "model.layers.5.mlp.switch_mlp.gate_proj": true,
+        "model.layers.5.mlp.switch_mlp.up_proj": true,
+        "model.layers.5.mlp.switch_mlp.down_proj": true,
+        "model.layers.5.mlp.switch_mlp.activation": false,
+        "model.layers.5.mlp.shared_expert.gate_proj": true,
+        "model.layers.5.mlp.shared_expert.down_proj": true,
+        "model.layers.5.mlp.shared_expert.up_proj": true,
+        "model.layers.5.mlp.shared_expert_gate": true,
+        "model.layers.5.input_layernorm": false,
+        "model.layers.5.post_attention_layernorm": false,
+        "model.layers.6.self_attn.q_proj": true,
+        "model.layers.6.self_attn.k_proj": true,
+        "model.layers.6.self_attn.v_proj": true,
+        "model.layers.6.self_attn.o_proj": true,
+        "model.layers.6.self_attn.rope": false,
+        "model.layers.6.mlp.gate": true,
+        "model.layers.6.mlp.switch_mlp.gate_proj": true,
+        "model.layers.6.mlp.switch_mlp.up_proj": true,
+        "model.layers.6.mlp.switch_mlp.down_proj": true,
+        "model.layers.6.mlp.switch_mlp.activation": false,
+        "model.layers.6.mlp.shared_expert.gate_proj": true,
+        "model.layers.6.mlp.shared_expert.down_proj": true,
+        "model.layers.6.mlp.shared_expert.up_proj": true,
+        "model.layers.6.mlp.shared_expert_gate": true,
+        "model.layers.6.input_layernorm": false,
+        "model.layers.6.post_attention_layernorm": false,
+        "model.layers.7.self_attn.q_proj": true,
+        "model.layers.7.self_attn.k_proj": true,
+        "model.layers.7.self_attn.v_proj": true,
+        "model.layers.7.self_attn.o_proj": true,
+        "model.layers.7.self_attn.rope": false,
+        "model.layers.7.mlp.gate": true,
+        "model.layers.7.mlp.switch_mlp.gate_proj": true,
+        "model.layers.7.mlp.switch_mlp.up_proj": true,
+        "model.layers.7.mlp.switch_mlp.down_proj": true,
+        "model.layers.7.mlp.switch_mlp.activation": false,
+        "model.layers.7.mlp.shared_expert.gate_proj": true,
+        "model.layers.7.mlp.shared_expert.down_proj": true,
+        "model.layers.7.mlp.shared_expert.up_proj": true,
+        "model.layers.7.mlp.shared_expert_gate": true,
+        "model.layers.7.input_layernorm": false,
+        "model.layers.7.post_attention_layernorm": false,
+        "model.layers.8.self_attn.q_proj": true,
+        "model.layers.8.self_attn.k_proj": true,
+        "model.layers.8.self_attn.v_proj": true,
+        "model.layers.8.self_attn.o_proj": true,
+        "model.layers.8.self_attn.rope": false,
+        "model.layers.8.mlp.gate": true,
+        "model.layers.8.mlp.switch_mlp.gate_proj": true,
+        "model.layers.8.mlp.switch_mlp.up_proj": true,
+        "model.layers.8.mlp.switch_mlp.down_proj": true,
+        "model.layers.8.mlp.switch_mlp.activation": false,
+        "model.layers.8.mlp.shared_expert.gate_proj": true,
+        "model.layers.8.mlp.shared_expert.down_proj": true,
+        "model.layers.8.mlp.shared_expert.up_proj": true,
+        "model.layers.8.mlp.shared_expert_gate": true,
+        "model.layers.8.input_layernorm": false,
+        "model.layers.8.post_attention_layernorm": false,
+        "model.layers.9.self_attn.q_proj": true,
+        "model.layers.9.self_attn.k_proj": true,
+        "model.layers.9.self_attn.v_proj": true,
+        "model.layers.9.self_attn.o_proj": true,
+        "model.layers.9.self_attn.rope": false,
+        "model.layers.9.mlp.gate": true,
+        "model.layers.9.mlp.switch_mlp.gate_proj": true,
+        "model.layers.9.mlp.switch_mlp.up_proj": true,
+        "model.layers.9.mlp.switch_mlp.down_proj": true,
+        "model.layers.9.mlp.switch_mlp.activation": false,
+        "model.layers.9.mlp.shared_expert.gate_proj": true,
+        "model.layers.9.mlp.shared_expert.down_proj": true,
+        "model.layers.9.mlp.shared_expert.up_proj": true,
+        "model.layers.9.mlp.shared_expert_gate": true,
+        "model.layers.9.input_layernorm": false,
+        "model.layers.9.post_attention_layernorm": false,
+        "model.layers.10.self_attn.q_proj": true,
+        "model.layers.10.self_attn.k_proj": true,
+        "model.layers.10.self_attn.v_proj": true,
+        "model.layers.10.self_attn.o_proj": true,
+        "model.layers.10.self_attn.rope": false,
+        "model.layers.10.mlp.gate": true,
+        "model.layers.10.mlp.switch_mlp.gate_proj": true,
+        "model.layers.10.mlp.switch_mlp.up_proj": true,
+        "model.layers.10.mlp.switch_mlp.down_proj": true,
+        "model.layers.10.mlp.switch_mlp.activation": false,
+        "model.layers.10.mlp.shared_expert.gate_proj": true,
+        "model.layers.10.mlp.shared_expert.down_proj": true,
+        "model.layers.10.mlp.shared_expert.up_proj": true,
+        "model.layers.10.mlp.shared_expert_gate": true,
+        "model.layers.10.input_layernorm": false,
+        "model.layers.10.post_attention_layernorm": false,
+        "model.layers.11.self_attn.q_proj": true,
+        "model.layers.11.self_attn.k_proj": true,
+        "model.layers.11.self_attn.v_proj": true,
+        "model.layers.11.self_attn.o_proj": true,
+        "model.layers.11.self_attn.rope": false,
+        "model.layers.11.mlp.gate": true,
+        "model.layers.11.mlp.switch_mlp.gate_proj": true,
+        "model.layers.11.mlp.switch_mlp.up_proj": true,
+        "model.layers.11.mlp.switch_mlp.down_proj": true,
+        "model.layers.11.mlp.switch_mlp.activation": false,
+        "model.layers.11.mlp.shared_expert.gate_proj": true,
+        "model.layers.11.mlp.shared_expert.down_proj": true,
+        "model.layers.11.mlp.shared_expert.up_proj": true,
+        "model.layers.11.mlp.shared_expert_gate": true,
+        "model.layers.11.input_layernorm": false,
+        "model.layers.11.post_attention_layernorm": false,
+        "model.layers.12.self_attn.q_proj": true,
+        "model.layers.12.self_attn.k_proj": true,
+        "model.layers.12.self_attn.v_proj": true,
+        "model.layers.12.self_attn.o_proj": true,
+        "model.layers.12.self_attn.rope": false,
+        "model.layers.12.mlp.gate": true,
+        "model.layers.12.mlp.switch_mlp.gate_proj": true,
+        "model.layers.12.mlp.switch_mlp.up_proj": true,
+        "model.layers.12.mlp.switch_mlp.down_proj": true,
+        "model.layers.12.mlp.switch_mlp.activation": false,
+        "model.layers.12.mlp.shared_expert.gate_proj": true,
+        "model.layers.12.mlp.shared_expert.down_proj": true,
+        "model.layers.12.mlp.shared_expert.up_proj": true,
+        "model.layers.12.mlp.shared_expert_gate": true,
+        "model.layers.12.input_layernorm": false,
+        "model.layers.12.post_attention_layernorm": false,
+        "model.layers.13.self_attn.q_proj": true,
+        "model.layers.13.self_attn.k_proj": true,
+        "model.layers.13.self_attn.v_proj": true,
+        "model.layers.13.self_attn.o_proj": true,
+        "model.layers.13.self_attn.rope": false,
+        "model.layers.13.mlp.gate": true,
+        "model.layers.13.mlp.switch_mlp.gate_proj": true,
+        "model.layers.13.mlp.switch_mlp.up_proj": true,
+        "model.layers.13.mlp.switch_mlp.down_proj": true,
+        "model.layers.13.mlp.switch_mlp.activation": false,
+        "model.layers.13.mlp.shared_expert.gate_proj": true,
+        "model.layers.13.mlp.shared_expert.down_proj": true,
+        "model.layers.13.mlp.shared_expert.up_proj": true,
+        "model.layers.13.mlp.shared_expert_gate": true,
+        "model.layers.13.input_layernorm": false,
+        "model.layers.13.post_attention_layernorm": false,
+        "model.layers.14.self_attn.q_proj": true,
+        "model.layers.14.self_attn.k_proj": true,
+        "model.layers.14.self_attn.v_proj": true,
+        "model.layers.14.self_attn.o_proj": true,
+        "model.layers.14.self_attn.rope": false,
+        "model.layers.14.mlp.gate": true,
+        "model.layers.14.mlp.switch_mlp.gate_proj": true,
+        "model.layers.14.mlp.switch_mlp.up_proj": true,
+        "model.layers.14.mlp.switch_mlp.down_proj": true,
+        "model.layers.14.mlp.switch_mlp.activation": false,
+        "model.layers.14.mlp.shared_expert.gate_proj": true,
+        "model.layers.14.mlp.shared_expert.down_proj": true,
+        "model.layers.14.mlp.shared_expert.up_proj": true,
+        "model.layers.14.mlp.shared_expert_gate": true,
+        "model.layers.14.input_layernorm": false,
+        "model.layers.14.post_attention_layernorm": false,
+        "model.layers.15.self_attn.q_proj": true,
+        "model.layers.15.self_attn.k_proj": true,
+        "model.layers.15.self_attn.v_proj": true,
+        "model.layers.15.self_attn.o_proj": true,
+        "model.layers.15.self_attn.rope": false,
+        "model.layers.15.mlp.gate": true,
+        "model.layers.15.mlp.switch_mlp.gate_proj": true,
+        "model.layers.15.mlp.switch_mlp.up_proj": true,
+        "model.layers.15.mlp.switch_mlp.down_proj": true,
+        "model.layers.15.mlp.switch_mlp.activation": false,
+        "model.layers.15.mlp.shared_expert.gate_proj": true,
+        "model.layers.15.mlp.shared_expert.down_proj": true,
+        "model.layers.15.mlp.shared_expert.up_proj": true,
+        "model.layers.15.mlp.shared_expert_gate": true,
+        "model.layers.15.input_layernorm": false,
+        "model.layers.15.post_attention_layernorm": false,
+        "model.layers.16.self_attn.q_proj": true,
+        "model.layers.16.self_attn.k_proj": true,
+        "model.layers.16.self_attn.v_proj": true,
+        "model.layers.16.self_attn.o_proj": true,
+        "model.layers.16.self_attn.rope": false,
+        "model.layers.16.mlp.gate": true,
+        "model.layers.16.mlp.switch_mlp.gate_proj": true,
+        "model.layers.16.mlp.switch_mlp.up_proj": true,
+        "model.layers.16.mlp.switch_mlp.down_proj": true,
+        "model.layers.16.mlp.switch_mlp.activation": false,
+        "model.layers.16.mlp.shared_expert.gate_proj": true,
+        "model.layers.16.mlp.shared_expert.down_proj": true,
+        "model.layers.16.mlp.shared_expert.up_proj": true,
+        "model.layers.16.mlp.shared_expert_gate": true,
+        "model.layers.16.input_layernorm": false,
+        "model.layers.16.post_attention_layernorm": false,
+        "model.layers.17.self_attn.q_proj": true,
+        "model.layers.17.self_attn.k_proj": true,
+        "model.layers.17.self_attn.v_proj": true,
+        "model.layers.17.self_attn.o_proj": true,
+        "model.layers.17.self_attn.rope": false,
+        "model.layers.17.mlp.gate": true,
+        "model.layers.17.mlp.switch_mlp.gate_proj": true,
+        "model.layers.17.mlp.switch_mlp.up_proj": true,
+        "model.layers.17.mlp.switch_mlp.down_proj": true,
+        "model.layers.17.mlp.switch_mlp.activation": false,
+        "model.layers.17.mlp.shared_expert.gate_proj": true,
+        "model.layers.17.mlp.shared_expert.down_proj": true,
+        "model.layers.17.mlp.shared_expert.up_proj": true,
+        "model.layers.17.mlp.shared_expert_gate": true,
+        "model.layers.17.input_layernorm": false,
+        "model.layers.17.post_attention_layernorm": false,
+        "model.layers.18.self_attn.q_proj": true,
+        "model.layers.18.self_attn.k_proj": true,
+        "model.layers.18.self_attn.v_proj": true,
+        "model.layers.18.self_attn.o_proj": true,
+        "model.layers.18.self_attn.rope": false,
+        "model.layers.18.mlp.gate": true,
+        "model.layers.18.mlp.switch_mlp.gate_proj": true,
+        "model.layers.18.mlp.switch_mlp.up_proj": true,
+        "model.layers.18.mlp.switch_mlp.down_proj": true,
+        "model.layers.18.mlp.switch_mlp.activation": false,
+        "model.layers.18.mlp.shared_expert.gate_proj": true,
+        "model.layers.18.mlp.shared_expert.down_proj": true,
+        "model.layers.18.mlp.shared_expert.up_proj": true,
+        "model.layers.18.mlp.shared_expert_gate": true,
+        "model.layers.18.input_layernorm": false,
+        "model.layers.18.post_attention_layernorm": false,
+        "model.layers.19.self_attn.q_proj": true,
+        "model.layers.19.self_attn.k_proj": true,
+        "model.layers.19.self_attn.v_proj": true,
+        "model.layers.19.self_attn.o_proj": true,
+        "model.layers.19.self_attn.rope": false,
+        "model.layers.19.mlp.gate": true,
+        "model.layers.19.mlp.switch_mlp.gate_proj": true,
+        "model.layers.19.mlp.switch_mlp.up_proj": true,
+        "model.layers.19.mlp.switch_mlp.down_proj": true,
+        "model.layers.19.mlp.switch_mlp.activation": false,
+        "model.layers.19.mlp.shared_expert.gate_proj": true,
+        "model.layers.19.mlp.shared_expert.down_proj": true,
+        "model.layers.19.mlp.shared_expert.up_proj": true,
+        "model.layers.19.mlp.shared_expert_gate": true,
+        "model.layers.19.input_layernorm": false,
+        "model.layers.19.post_attention_layernorm": false,
+        "model.layers.20.self_attn.q_proj": true,
+        "model.layers.20.self_attn.k_proj": true,
+        "model.layers.20.self_attn.v_proj": true,
+        "model.layers.20.self_attn.o_proj": true,
+        "model.layers.20.self_attn.rope": false,
+        "model.layers.20.mlp.gate": true,
+        "model.layers.20.mlp.switch_mlp.gate_proj": true,
+        "model.layers.20.mlp.switch_mlp.up_proj": true,
+        "model.layers.20.mlp.switch_mlp.down_proj": true,
+        "model.layers.20.mlp.switch_mlp.activation": false,
+        "model.layers.20.mlp.shared_expert.gate_proj": true,
+        "model.layers.20.mlp.shared_expert.down_proj": true,
+        "model.layers.20.mlp.shared_expert.up_proj": true,
+        "model.layers.20.mlp.shared_expert_gate": true,
+        "model.layers.20.input_layernorm": false,
+        "model.layers.20.post_attention_layernorm": false,
+        "model.layers.21.self_attn.q_proj": true,
+        "model.layers.21.self_attn.k_proj": true,
+        "model.layers.21.self_attn.v_proj": true,
+        "model.layers.21.self_attn.o_proj": true,
+        "model.layers.21.self_attn.rope": false,
+        "model.layers.21.mlp.gate": true,
+        "model.layers.21.mlp.switch_mlp.gate_proj": true,
+        "model.layers.21.mlp.switch_mlp.up_proj": true,
+        "model.layers.21.mlp.switch_mlp.down_proj": true,
+        "model.layers.21.mlp.switch_mlp.activation": false,
+        "model.layers.21.mlp.shared_expert.gate_proj": true,
+        "model.layers.21.mlp.shared_expert.down_proj": true,
+        "model.layers.21.mlp.shared_expert.up_proj": true,
+        "model.layers.21.mlp.shared_expert_gate": true,
+        "model.layers.21.input_layernorm": false,
+        "model.layers.21.post_attention_layernorm": false,
+        "model.layers.22.self_attn.q_proj": true,
+        "model.layers.22.self_attn.k_proj": true,
+        "model.layers.22.self_attn.v_proj": true,
+        "model.layers.22.self_attn.o_proj": true,
+        "model.layers.22.self_attn.rope": false,
+        "model.layers.22.mlp.gate": true,
+        "model.layers.22.mlp.switch_mlp.gate_proj": true,
+        "model.layers.22.mlp.switch_mlp.up_proj": true,
+        "model.layers.22.mlp.switch_mlp.down_proj": true,
+        "model.layers.22.mlp.switch_mlp.activation": false,
+        "model.layers.22.mlp.shared_expert.gate_proj": true,
+        "model.layers.22.mlp.shared_expert.down_proj": true,
+        "model.layers.22.mlp.shared_expert.up_proj": true,
+        "model.layers.22.mlp.shared_expert_gate": true,
+        "model.layers.22.input_layernorm": false,
+        "model.layers.22.post_attention_layernorm": false,
+        "model.layers.23.self_attn.q_proj": true,
+        "model.layers.23.self_attn.k_proj": true,
+        "model.layers.23.self_attn.v_proj": true,
+        "model.layers.23.self_attn.o_proj": true,
+        "model.layers.23.self_attn.rope": false,
+        "model.layers.23.mlp.gate": true,
+        "model.layers.23.mlp.switch_mlp.gate_proj": true,
+        "model.layers.23.mlp.switch_mlp.up_proj": true,
+        "model.layers.23.mlp.switch_mlp.down_proj": true,
+        "model.layers.23.mlp.switch_mlp.activation": false,
+        "model.layers.23.mlp.shared_expert.gate_proj": true,
+        "model.layers.23.mlp.shared_expert.down_proj": true,
+        "model.layers.23.mlp.shared_expert.up_proj": true,
+        "model.layers.23.mlp.shared_expert_gate": true,
+        "model.layers.23.input_layernorm": false,
+        "model.layers.23.post_attention_layernorm": false,
+        "model.layers.24.self_attn.q_proj": true,
+        "model.layers.24.self_attn.k_proj": true,
+        "model.layers.24.self_attn.v_proj": true,
+        "model.layers.24.self_attn.o_proj": true,
+        "model.layers.24.self_attn.rope": false,
+        "model.layers.24.mlp.gate": true,
+        "model.layers.24.mlp.switch_mlp.gate_proj": true,
+        "model.layers.24.mlp.switch_mlp.up_proj": true,
+        "model.layers.24.mlp.switch_mlp.down_proj": true,
+        "model.layers.24.mlp.switch_mlp.activation": false,
+        "model.layers.24.mlp.shared_expert.gate_proj": true,
+        "model.layers.24.mlp.shared_expert.down_proj": true,
+        "model.layers.24.mlp.shared_expert.up_proj": true,
+        "model.layers.24.mlp.shared_expert_gate": true,
+        "model.layers.24.input_layernorm": false,
+        "model.layers.24.post_attention_layernorm": false,
+        "model.layers.25.self_attn.q_proj": true,
+        "model.layers.25.self_attn.k_proj": true,
+        "model.layers.25.self_attn.v_proj": true,
+        "model.layers.25.self_attn.o_proj": true,
+        "model.layers.25.self_attn.rope": false,
+        "model.layers.25.mlp.gate": true,
+        "model.layers.25.mlp.switch_mlp.gate_proj": true,
+        "model.layers.25.mlp.switch_mlp.up_proj": true,
+        "model.layers.25.mlp.switch_mlp.down_proj": true,
+        "model.layers.25.mlp.switch_mlp.activation": false,
+        "model.layers.25.mlp.shared_expert.gate_proj": true,
+        "model.layers.25.mlp.shared_expert.down_proj": true,
+        "model.layers.25.mlp.shared_expert.up_proj": true,
+        "model.layers.25.mlp.shared_expert_gate": true,
+        "model.layers.25.input_layernorm": false,
+        "model.layers.25.post_attention_layernorm": false,
+        "model.layers.26.self_attn.q_proj": true,
+        "model.layers.26.self_attn.k_proj": true,
+        "model.layers.26.self_attn.v_proj": true,
+        "model.layers.26.self_attn.o_proj": true,
+        "model.layers.26.self_attn.rope": false,
+        "model.layers.26.mlp.gate": true,
+        "model.layers.26.mlp.switch_mlp.gate_proj": true,
+        "model.layers.26.mlp.switch_mlp.up_proj": true,
+        "model.layers.26.mlp.switch_mlp.down_proj": true,
+        "model.layers.26.mlp.switch_mlp.activation": false,
+        "model.layers.26.mlp.shared_expert.gate_proj": true,
+        "model.layers.26.mlp.shared_expert.down_proj": true,
+        "model.layers.26.mlp.shared_expert.up_proj": true,
+        "model.layers.26.mlp.shared_expert_gate": true,
+        "model.layers.26.input_layernorm": false,
+        "model.layers.26.post_attention_layernorm": false,
+        "model.layers.27.self_attn.q_proj": true,
+        "model.layers.27.self_attn.k_proj": true,
+        "model.layers.27.self_attn.v_proj": true,
+        "model.layers.27.self_attn.o_proj": true,
+        "model.layers.27.self_attn.rope": false,
+        "model.layers.27.mlp.gate": {
+            "bits": 8,
+            "group_size": 32
+        },
+        "model.layers.27.mlp.switch_mlp.gate_proj": true,
+        "model.layers.27.mlp.switch_mlp.up_proj": true,
+        "model.layers.27.mlp.switch_mlp.down_proj": true,
+        "model.layers.27.mlp.switch_mlp.activation": false,
+        "model.layers.27.mlp.shared_expert.gate_proj": true,
+        "model.layers.27.mlp.shared_expert.down_proj": true,
+        "model.layers.27.mlp.shared_expert.up_proj": true,
+        "model.layers.27.mlp.shared_expert_gate": true,
+        "model.layers.27.input_layernorm": false,
+        "model.layers.27.post_attention_layernorm": false,
+        "model.norm": false,
+        "lm_head": true
+    },
+    "quantization_config": {
+        "group_size": 32,
+        "bits": 6,
+        "model.embed_tokens": true,
+        "model.layers.0.self_attn.q_proj": true,
+        "model.layers.0.self_attn.k_proj": true,
+        "model.layers.0.self_attn.v_proj": true,
+        "model.layers.0.self_attn.o_proj": true,
+        "model.layers.0.self_attn.rope": false,
+        "model.layers.0.mlp.gate": true,
+        "model.layers.0.mlp.switch_mlp.gate_proj": true,
+        "model.layers.0.mlp.switch_mlp.up_proj": true,
+        "model.layers.0.mlp.switch_mlp.down_proj": true,
+        "model.layers.0.mlp.switch_mlp.activation": false,
+        "model.layers.0.mlp.shared_expert.gate_proj": true,
+        "model.layers.0.mlp.shared_expert.down_proj": true,
+        "model.layers.0.mlp.shared_expert.up_proj": true,
+        "model.layers.0.mlp.shared_expert_gate": true,
+        "model.layers.0.input_layernorm": false,
+        "model.layers.0.post_attention_layernorm": false,
+        "model.layers.1.self_attn.q_proj": true,
+        "model.layers.1.self_attn.k_proj": true,
+        "model.layers.1.self_attn.v_proj": true,
+        "model.layers.1.self_attn.o_proj": true,
+        "model.layers.1.self_attn.rope": false,
+        "model.layers.1.mlp.gate": true,
+        "model.layers.1.mlp.switch_mlp.gate_proj": true,
+        "model.layers.1.mlp.switch_mlp.up_proj": true,
+        "model.layers.1.mlp.switch_mlp.down_proj": true,
+        "model.layers.1.mlp.switch_mlp.activation": false,
+        "model.layers.1.mlp.shared_expert.gate_proj": true,
+        "model.layers.1.mlp.shared_expert.down_proj": true,
+        "model.layers.1.mlp.shared_expert.up_proj": true,
+        "model.layers.1.mlp.shared_expert_gate": true,
+        "model.layers.1.input_layernorm": false,
+        "model.layers.1.post_attention_layernorm": false,
+        "model.layers.2.self_attn.q_proj": true,
+        "model.layers.2.self_attn.k_proj": true,
+        "model.layers.2.self_attn.v_proj": true,
+        "model.layers.2.self_attn.o_proj": true,
+        "model.layers.2.self_attn.rope": false,
+        "model.layers.2.mlp.gate": true,
+        "model.layers.2.mlp.switch_mlp.gate_proj": true,
+        "model.layers.2.mlp.switch_mlp.up_proj": true,
+        "model.layers.2.mlp.switch_mlp.down_proj": true,
+        "model.layers.2.mlp.switch_mlp.activation": false,
+        "model.layers.2.mlp.shared_expert.gate_proj": true,
+        "model.layers.2.mlp.shared_expert.down_proj": true,
+        "model.layers.2.mlp.shared_expert.up_proj": true,
+        "model.layers.2.mlp.shared_expert_gate": true,
+        "model.layers.2.input_layernorm": false,
+        "model.layers.2.post_attention_layernorm": false,
+        "model.layers.3.self_attn.q_proj": true,
+        "model.layers.3.self_attn.k_proj": true,
+        "model.layers.3.self_attn.v_proj": true,
+        "model.layers.3.self_attn.o_proj": true,
+        "model.layers.3.self_attn.rope": false,
+        "model.layers.3.mlp.gate": true,
+        "model.layers.3.mlp.switch_mlp.gate_proj": true,
+        "model.layers.3.mlp.switch_mlp.up_proj": true,
+        "model.layers.3.mlp.switch_mlp.down_proj": true,
+        "model.layers.3.mlp.switch_mlp.activation": false,
+        "model.layers.3.mlp.shared_expert.gate_proj": true,
+        "model.layers.3.mlp.shared_expert.down_proj": true,
+        "model.layers.3.mlp.shared_expert.up_proj": true,
+        "model.layers.3.mlp.shared_expert_gate": true,
+        "model.layers.3.input_layernorm": false,
+        "model.layers.3.post_attention_layernorm": false,
+        "model.layers.4.self_attn.q_proj": true,
+        "model.layers.4.self_attn.k_proj": true,
+        "model.layers.4.self_attn.v_proj": true,
+        "model.layers.4.self_attn.o_proj": true,
+        "model.layers.4.self_attn.rope": false,
+        "model.layers.4.mlp.gate": true,
+        "model.layers.4.mlp.switch_mlp.gate_proj": true,
+        "model.layers.4.mlp.switch_mlp.up_proj": true,
+        "model.layers.4.mlp.switch_mlp.down_proj": true,
+        "model.layers.4.mlp.switch_mlp.activation": false,
+        "model.layers.4.mlp.shared_expert.gate_proj": true,
+        "model.layers.4.mlp.shared_expert.down_proj": true,
+        "model.layers.4.mlp.shared_expert.up_proj": true,
+        "model.layers.4.mlp.shared_expert_gate": true,
+        "model.layers.4.input_layernorm": false,
+        "model.layers.4.post_attention_layernorm": false,
+        "model.layers.5.self_attn.q_proj": true,
+        "model.layers.5.self_attn.k_proj": true,
+        "model.layers.5.self_attn.v_proj": true,
+        "model.layers.5.self_attn.o_proj": true,
+        "model.layers.5.self_attn.rope": false,
+        "model.layers.5.mlp.gate": true,
+        "model.layers.5.mlp.switch_mlp.gate_proj": true,
+        "model.layers.5.mlp.switch_mlp.up_proj": true,
+        "model.layers.5.mlp.switch_mlp.down_proj": true,
+        "model.layers.5.mlp.switch_mlp.activation": false,
+        "model.layers.5.mlp.shared_expert.gate_proj": true,
+        "model.layers.5.mlp.shared_expert.down_proj": true,
+        "model.layers.5.mlp.shared_expert.up_proj": true,
+        "model.layers.5.mlp.shared_expert_gate": true,
+        "model.layers.5.input_layernorm": false,
+        "model.layers.5.post_attention_layernorm": false,
+        "model.layers.6.self_attn.q_proj": true,
+        "model.layers.6.self_attn.k_proj": true,
+        "model.layers.6.self_attn.v_proj": true,
+        "model.layers.6.self_attn.o_proj": true,
+        "model.layers.6.self_attn.rope": false,
+        "model.layers.6.mlp.gate": true,
+        "model.layers.6.mlp.switch_mlp.gate_proj": true,
+        "model.layers.6.mlp.switch_mlp.up_proj": true,
+        "model.layers.6.mlp.switch_mlp.down_proj": true,
+        "model.layers.6.mlp.switch_mlp.activation": false,
+        "model.layers.6.mlp.shared_expert.gate_proj": true,
+        "model.layers.6.mlp.shared_expert.down_proj": true,
+        "model.layers.6.mlp.shared_expert.up_proj": true,
+        "model.layers.6.mlp.shared_expert_gate": true,
+        "model.layers.6.input_layernorm": false,
+        "model.layers.6.post_attention_layernorm": false,
+        "model.layers.7.self_attn.q_proj": true,
+        "model.layers.7.self_attn.k_proj": true,
+        "model.layers.7.self_attn.v_proj": true,
+        "model.layers.7.self_attn.o_proj": true,
+        "model.layers.7.self_attn.rope": false,
+        "model.layers.7.mlp.gate": true,
+        "model.layers.7.mlp.switch_mlp.gate_proj": true,
+        "model.layers.7.mlp.switch_mlp.up_proj": true,
+        "model.layers.7.mlp.switch_mlp.down_proj": true,
+        "model.layers.7.mlp.switch_mlp.activation": false,
+        "model.layers.7.mlp.shared_expert.gate_proj": true,
+        "model.layers.7.mlp.shared_expert.down_proj": true,
+        "model.layers.7.mlp.shared_expert.up_proj": true,
+        "model.layers.7.mlp.shared_expert_gate": true,
+        "model.layers.7.input_layernorm": false,
+        "model.layers.7.post_attention_layernorm": false,
+        "model.layers.8.self_attn.q_proj": true,
+        "model.layers.8.self_attn.k_proj": true,
+        "model.layers.8.self_attn.v_proj": true,
+        "model.layers.8.self_attn.o_proj": true,
+        "model.layers.8.self_attn.rope": false,
+        "model.layers.8.mlp.gate": true,
+        "model.layers.8.mlp.switch_mlp.gate_proj": true,
+        "model.layers.8.mlp.switch_mlp.up_proj": true,
+        "model.layers.8.mlp.switch_mlp.down_proj": true,
+        "model.layers.8.mlp.switch_mlp.activation": false,
+        "model.layers.8.mlp.shared_expert.gate_proj": true,
+        "model.layers.8.mlp.shared_expert.down_proj": true,
+        "model.layers.8.mlp.shared_expert.up_proj": true,
+        "model.layers.8.mlp.shared_expert_gate": true,
+        "model.layers.8.input_layernorm": false,
+        "model.layers.8.post_attention_layernorm": false,
+        "model.layers.9.self_attn.q_proj": true,
+        "model.layers.9.self_attn.k_proj": true,
+        "model.layers.9.self_attn.v_proj": true,
+        "model.layers.9.self_attn.o_proj": true,
+        "model.layers.9.self_attn.rope": false,
+        "model.layers.9.mlp.gate": true,
+        "model.layers.9.mlp.switch_mlp.gate_proj": true,
+        "model.layers.9.mlp.switch_mlp.up_proj": true,
+        "model.layers.9.mlp.switch_mlp.down_proj": true,
+        "model.layers.9.mlp.switch_mlp.activation": false,
+        "model.layers.9.mlp.shared_expert.gate_proj": true,
+        "model.layers.9.mlp.shared_expert.down_proj": true,
+        "model.layers.9.mlp.shared_expert.up_proj": true,
+        "model.layers.9.mlp.shared_expert_gate": true,
+        "model.layers.9.input_layernorm": false,
+        "model.layers.9.post_attention_layernorm": false,
+        "model.layers.10.self_attn.q_proj": true,
+        "model.layers.10.self_attn.k_proj": true,
+        "model.layers.10.self_attn.v_proj": true,
+        "model.layers.10.self_attn.o_proj": true,
+        "model.layers.10.self_attn.rope": false,
+        "model.layers.10.mlp.gate": true,
+        "model.layers.10.mlp.switch_mlp.gate_proj": true,
+        "model.layers.10.mlp.switch_mlp.up_proj": true,
+        "model.layers.10.mlp.switch_mlp.down_proj": true,
+        "model.layers.10.mlp.switch_mlp.activation": false,
+        "model.layers.10.mlp.shared_expert.gate_proj": true,
+        "model.layers.10.mlp.shared_expert.down_proj": true,
+        "model.layers.10.mlp.shared_expert.up_proj": true,
+        "model.layers.10.mlp.shared_expert_gate": true,
+        "model.layers.10.input_layernorm": false,
+        "model.layers.10.post_attention_layernorm": false,
+        "model.layers.11.self_attn.q_proj": true,
+        "model.layers.11.self_attn.k_proj": true,
+        "model.layers.11.self_attn.v_proj": true,
+        "model.layers.11.self_attn.o_proj": true,
+        "model.layers.11.self_attn.rope": false,
+        "model.layers.11.mlp.gate": true,
+        "model.layers.11.mlp.switch_mlp.gate_proj": true,
+        "model.layers.11.mlp.switch_mlp.up_proj": true,
+        "model.layers.11.mlp.switch_mlp.down_proj": true,
+        "model.layers.11.mlp.switch_mlp.activation": false,
+        "model.layers.11.mlp.shared_expert.gate_proj": true,
+        "model.layers.11.mlp.shared_expert.down_proj": true,
+        "model.layers.11.mlp.shared_expert.up_proj": true,
+        "model.layers.11.mlp.shared_expert_gate": true,
+        "model.layers.11.input_layernorm": false,
+        "model.layers.11.post_attention_layernorm": false,
+        "model.layers.12.self_attn.q_proj": true,
+        "model.layers.12.self_attn.k_proj": true,
+        "model.layers.12.self_attn.v_proj": true,
+        "model.layers.12.self_attn.o_proj": true,
+        "model.layers.12.self_attn.rope": false,
+        "model.layers.12.mlp.gate": true,
+        "model.layers.12.mlp.switch_mlp.gate_proj": true,
+        "model.layers.12.mlp.switch_mlp.up_proj": true,
+        "model.layers.12.mlp.switch_mlp.down_proj": true,
+        "model.layers.12.mlp.switch_mlp.activation": false,
+        "model.layers.12.mlp.shared_expert.gate_proj": true,
+        "model.layers.12.mlp.shared_expert.down_proj": true,
+        "model.layers.12.mlp.shared_expert.up_proj": true,
+        "model.layers.12.mlp.shared_expert_gate": true,
+        "model.layers.12.input_layernorm": false,
+        "model.layers.12.post_attention_layernorm": false,
+        "model.layers.13.self_attn.q_proj": true,
+        "model.layers.13.self_attn.k_proj": true,
+        "model.layers.13.self_attn.v_proj": true,
+        "model.layers.13.self_attn.o_proj": true,
+        "model.layers.13.self_attn.rope": false,
+        "model.layers.13.mlp.gate": true,
+        "model.layers.13.mlp.switch_mlp.gate_proj": true,
+        "model.layers.13.mlp.switch_mlp.up_proj": true,
+        "model.layers.13.mlp.switch_mlp.down_proj": true,
+        "model.layers.13.mlp.switch_mlp.activation": false,
+        "model.layers.13.mlp.shared_expert.gate_proj": true,
+        "model.layers.13.mlp.shared_expert.down_proj": true,
+        "model.layers.13.mlp.shared_expert.up_proj": true,
+        "model.layers.13.mlp.shared_expert_gate": true,
+        "model.layers.13.input_layernorm": false,
+        "model.layers.13.post_attention_layernorm": false,
+        "model.layers.14.self_attn.q_proj": true,
+        "model.layers.14.self_attn.k_proj": true,
+        "model.layers.14.self_attn.v_proj": true,
+        "model.layers.14.self_attn.o_proj": true,
+        "model.layers.14.self_attn.rope": false,
+        "model.layers.14.mlp.gate": true,
+        "model.layers.14.mlp.switch_mlp.gate_proj": true,
+        "model.layers.14.mlp.switch_mlp.up_proj": true,
+        "model.layers.14.mlp.switch_mlp.down_proj": true,
+        "model.layers.14.mlp.switch_mlp.activation": false,
+        "model.layers.14.mlp.shared_expert.gate_proj": true,
+        "model.layers.14.mlp.shared_expert.down_proj": true,
+        "model.layers.14.mlp.shared_expert.up_proj": true,
+        "model.layers.14.mlp.shared_expert_gate": true,
+        "model.layers.14.input_layernorm": false,
+        "model.layers.14.post_attention_layernorm": false,
+        "model.layers.15.self_attn.q_proj": true,
+        "model.layers.15.self_attn.k_proj": true,
+        "model.layers.15.self_attn.v_proj": true,
+        "model.layers.15.self_attn.o_proj": true,
+        "model.layers.15.self_attn.rope": false,
+        "model.layers.15.mlp.gate": true,
+        "model.layers.15.mlp.switch_mlp.gate_proj": true,
+        "model.layers.15.mlp.switch_mlp.up_proj": true,
+        "model.layers.15.mlp.switch_mlp.down_proj": true,
+        "model.layers.15.mlp.switch_mlp.activation": false,
+        "model.layers.15.mlp.shared_expert.gate_proj": true,
+        "model.layers.15.mlp.shared_expert.down_proj": true,
+        "model.layers.15.mlp.shared_expert.up_proj": true,
+        "model.layers.15.mlp.shared_expert_gate": true,
+        "model.layers.15.input_layernorm": false,
+        "model.layers.15.post_attention_layernorm": false,
+        "model.layers.16.self_attn.q_proj": true,
+        "model.layers.16.self_attn.k_proj": true,
+        "model.layers.16.self_attn.v_proj": true,
+        "model.layers.16.self_attn.o_proj": true,
+        "model.layers.16.self_attn.rope": false,
+        "model.layers.16.mlp.gate": true,
+        "model.layers.16.mlp.switch_mlp.gate_proj": true,
+        "model.layers.16.mlp.switch_mlp.up_proj": true,
+        "model.layers.16.mlp.switch_mlp.down_proj": true,
+        "model.layers.16.mlp.switch_mlp.activation": false,
+        "model.layers.16.mlp.shared_expert.gate_proj": true,
+        "model.layers.16.mlp.shared_expert.down_proj": true,
+        "model.layers.16.mlp.shared_expert.up_proj": true,
+        "model.layers.16.mlp.shared_expert_gate": true,
+        "model.layers.16.input_layernorm": false,
+        "model.layers.16.post_attention_layernorm": false,
+        "model.layers.17.self_attn.q_proj": true,
+        "model.layers.17.self_attn.k_proj": true,
+        "model.layers.17.self_attn.v_proj": true,
+        "model.layers.17.self_attn.o_proj": true,
+        "model.layers.17.self_attn.rope": false,
+        "model.layers.17.mlp.gate": true,
+        "model.layers.17.mlp.switch_mlp.gate_proj": true,
+        "model.layers.17.mlp.switch_mlp.up_proj": true,
+        "model.layers.17.mlp.switch_mlp.down_proj": true,
+        "model.layers.17.mlp.switch_mlp.activation": false,
+        "model.layers.17.mlp.shared_expert.gate_proj": true,
+        "model.layers.17.mlp.shared_expert.down_proj": true,
+        "model.layers.17.mlp.shared_expert.up_proj": true,
+        "model.layers.17.mlp.shared_expert_gate": true,
+        "model.layers.17.input_layernorm": false,
+        "model.layers.17.post_attention_layernorm": false,
+        "model.layers.18.self_attn.q_proj": true,
+        "model.layers.18.self_attn.k_proj": true,
+        "model.layers.18.self_attn.v_proj": true,
+        "model.layers.18.self_attn.o_proj": true,
+        "model.layers.18.self_attn.rope": false,
+        "model.layers.18.mlp.gate": true,
+        "model.layers.18.mlp.switch_mlp.gate_proj": true,
+        "model.layers.18.mlp.switch_mlp.up_proj": true,
+        "model.layers.18.mlp.switch_mlp.down_proj": true,
+        "model.layers.18.mlp.switch_mlp.activation": false,
+        "model.layers.18.mlp.shared_expert.gate_proj": true,
+        "model.layers.18.mlp.shared_expert.down_proj": true,
+        "model.layers.18.mlp.shared_expert.up_proj": true,
+        "model.layers.18.mlp.shared_expert_gate": true,
+        "model.layers.18.input_layernorm": false,
+        "model.layers.18.post_attention_layernorm": false,
+        "model.layers.19.self_attn.q_proj": true,
+        "model.layers.19.self_attn.k_proj": true,
+        "model.layers.19.self_attn.v_proj": true,
+        "model.layers.19.self_attn.o_proj": true,
+        "model.layers.19.self_attn.rope": false,
+        "model.layers.19.mlp.gate": true,
+        "model.layers.19.mlp.switch_mlp.gate_proj": true,
+        "model.layers.19.mlp.switch_mlp.up_proj": true,
+        "model.layers.19.mlp.switch_mlp.down_proj": true,
+        "model.layers.19.mlp.switch_mlp.activation": false,
+        "model.layers.19.mlp.shared_expert.gate_proj": true,
+        "model.layers.19.mlp.shared_expert.down_proj": true,
+        "model.layers.19.mlp.shared_expert.up_proj": true,
+        "model.layers.19.mlp.shared_expert_gate": true,
+        "model.layers.19.input_layernorm": false,
+        "model.layers.19.post_attention_layernorm": false,
+        "model.layers.20.self_attn.q_proj": true,
+        "model.layers.20.self_attn.k_proj": true,
+        "model.layers.20.self_attn.v_proj": true,
+        "model.layers.20.self_attn.o_proj": true,
+        "model.layers.20.self_attn.rope": false,
+        "model.layers.20.mlp.gate": true,
+        "model.layers.20.mlp.switch_mlp.gate_proj": true,
+        "model.layers.20.mlp.switch_mlp.up_proj": true,
+        "model.layers.20.mlp.switch_mlp.down_proj": true,
+        "model.layers.20.mlp.switch_mlp.activation": false,
+        "model.layers.20.mlp.shared_expert.gate_proj": true,
+        "model.layers.20.mlp.shared_expert.down_proj": true,
+        "model.layers.20.mlp.shared_expert.up_proj": true,
+        "model.layers.20.mlp.shared_expert_gate": true,
+        "model.layers.20.input_layernorm": false,
+        "model.layers.20.post_attention_layernorm": false,
+        "model.layers.21.self_attn.q_proj": true,
+        "model.layers.21.self_attn.k_proj": true,
+        "model.layers.21.self_attn.v_proj": true,
+        "model.layers.21.self_attn.o_proj": true,
+        "model.layers.21.self_attn.rope": false,
+        "model.layers.21.mlp.gate": true,
+        "model.layers.21.mlp.switch_mlp.gate_proj": true,
+        "model.layers.21.mlp.switch_mlp.up_proj": true,
+        "model.layers.21.mlp.switch_mlp.down_proj": true,
+        "model.layers.21.mlp.switch_mlp.activation": false,
+        "model.layers.21.mlp.shared_expert.gate_proj": true,
+        "model.layers.21.mlp.shared_expert.down_proj": true,
+        "model.layers.21.mlp.shared_expert.up_proj": true,
+        "model.layers.21.mlp.shared_expert_gate": true,
+        "model.layers.21.input_layernorm": false,
+        "model.layers.21.post_attention_layernorm": false,
+        "model.layers.22.self_attn.q_proj": true,
+        "model.layers.22.self_attn.k_proj": true,
+        "model.layers.22.self_attn.v_proj": true,
+        "model.layers.22.self_attn.o_proj": true,
+        "model.layers.22.self_attn.rope": false,
+        "model.layers.22.mlp.gate": true,
+        "model.layers.22.mlp.switch_mlp.gate_proj": true,
+        "model.layers.22.mlp.switch_mlp.up_proj": true,
+        "model.layers.22.mlp.switch_mlp.down_proj": true,
+        "model.layers.22.mlp.switch_mlp.activation": false,
+        "model.layers.22.mlp.shared_expert.gate_proj": true,
+        "model.layers.22.mlp.shared_expert.down_proj": true,
+        "model.layers.22.mlp.shared_expert.up_proj": true,
+        "model.layers.22.mlp.shared_expert_gate": true,
+        "model.layers.22.input_layernorm": false,
+        "model.layers.22.post_attention_layernorm": false,
+        "model.layers.23.self_attn.q_proj": true,
+        "model.layers.23.self_attn.k_proj": true,
+        "model.layers.23.self_attn.v_proj": true,
+        "model.layers.23.self_attn.o_proj": true,
+        "model.layers.23.self_attn.rope": false,
+        "model.layers.23.mlp.gate": true,
+        "model.layers.23.mlp.switch_mlp.gate_proj": true,
+        "model.layers.23.mlp.switch_mlp.up_proj": true,
+        "model.layers.23.mlp.switch_mlp.down_proj": true,
+        "model.layers.23.mlp.switch_mlp.activation": false,
+        "model.layers.23.mlp.shared_expert.gate_proj": true,
+        "model.layers.23.mlp.shared_expert.down_proj": true,
+        "model.layers.23.mlp.shared_expert.up_proj": true,
+        "model.layers.23.mlp.shared_expert_gate": true,
+        "model.layers.23.input_layernorm": false,
+        "model.layers.23.post_attention_layernorm": false,
+        "model.layers.24.self_attn.q_proj": true,
+        "model.layers.24.self_attn.k_proj": true,
+        "model.layers.24.self_attn.v_proj": true,
+        "model.layers.24.self_attn.o_proj": true,
+        "model.layers.24.self_attn.rope": false,
+        "model.layers.24.mlp.gate": true,
+        "model.layers.24.mlp.switch_mlp.gate_proj": true,
+        "model.layers.24.mlp.switch_mlp.up_proj": true,
+        "model.layers.24.mlp.switch_mlp.down_proj": true,
+        "model.layers.24.mlp.switch_mlp.activation": false,
+        "model.layers.24.mlp.shared_expert.gate_proj": true,
+        "model.layers.24.mlp.shared_expert.down_proj": true,
+        "model.layers.24.mlp.shared_expert.up_proj": true,
+        "model.layers.24.mlp.shared_expert_gate": true,
+        "model.layers.24.input_layernorm": false,
+        "model.layers.24.post_attention_layernorm": false,
+        "model.layers.25.self_attn.q_proj": true,
+        "model.layers.25.self_attn.k_proj": true,
+        "model.layers.25.self_attn.v_proj": true,
+        "model.layers.25.self_attn.o_proj": true,
+        "model.layers.25.self_attn.rope": false,
+        "model.layers.25.mlp.gate": true,
+        "model.layers.25.mlp.switch_mlp.gate_proj": true,
+        "model.layers.25.mlp.switch_mlp.up_proj": true,
+        "model.layers.25.mlp.switch_mlp.down_proj": true,
+        "model.layers.25.mlp.switch_mlp.activation": false,
+        "model.layers.25.mlp.shared_expert.gate_proj": true,
+        "model.layers.25.mlp.shared_expert.down_proj": true,
+        "model.layers.25.mlp.shared_expert.up_proj": true,
+        "model.layers.25.mlp.shared_expert_gate": true,
+        "model.layers.25.input_layernorm": false,
+        "model.layers.25.post_attention_layernorm": false,
+        "model.layers.26.self_attn.q_proj": true,
+        "model.layers.26.self_attn.k_proj": true,
+        "model.layers.26.self_attn.v_proj": true,
+        "model.layers.26.self_attn.o_proj": true,
+        "model.layers.26.self_attn.rope": false,
+        "model.layers.26.mlp.gate": true,
+        "model.layers.26.mlp.switch_mlp.gate_proj": true,
+        "model.layers.26.mlp.switch_mlp.up_proj": true,
+        "model.layers.26.mlp.switch_mlp.down_proj": true,
+        "model.layers.26.mlp.switch_mlp.activation": false,
+        "model.layers.26.mlp.shared_expert.gate_proj": true,
+        "model.layers.26.mlp.shared_expert.down_proj": true,
+        "model.layers.26.mlp.shared_expert.up_proj": true,
+        "model.layers.26.mlp.shared_expert_gate": true,
+        "model.layers.26.input_layernorm": false,
+        "model.layers.26.post_attention_layernorm": false,
+        "model.layers.27.self_attn.q_proj": true,
+        "model.layers.27.self_attn.k_proj": true,
+        "model.layers.27.self_attn.v_proj": true,
+        "model.layers.27.self_attn.o_proj": true,
+        "model.layers.27.self_attn.rope": false,
+        "model.layers.27.mlp.gate": {
+            "bits": 8,
+            "group_size": 32
+        },
+        "model.layers.27.mlp.switch_mlp.gate_proj": true,
+        "model.layers.27.mlp.switch_mlp.up_proj": true,
+        "model.layers.27.mlp.switch_mlp.down_proj": true,
+        "model.layers.27.mlp.switch_mlp.activation": false,
+        "model.layers.27.mlp.shared_expert.gate_proj": true,
+        "model.layers.27.mlp.shared_expert.down_proj": true,
+        "model.layers.27.mlp.shared_expert.up_proj": true,
+        "model.layers.27.mlp.shared_expert_gate": true,
+        "model.layers.27.input_layernorm": false,
+        "model.layers.27.post_attention_layernorm": false,
+        "model.norm": false,
+        "lm_head": true
+    },
+    "resid_pdrop": 0.0,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 1000000.0,
+    "router_aux_loss_coef": 0.001,
+    "shared_expert_intermediate_size": 18944,
+    "sliding_window": null,
+    "tie_word_embeddings": false,
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.52.0.dev0",
+    "use_cache": false,
+    "use_sliding_window": false,
+    "vocab_size": 152064
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0471407a9d561944152bf1819d2f9c03fa5e29d92b8b99379687362607e21c0a
+size 5342259216

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a05fdb1e2e46d8a67cf9c3751cf0532cbaa2e001409b1871f5a480f5bd2a7653
+size 5366400061

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddda968643112f060d20b4c726a67ada2e29f11edd8c60f58c3c165d7874b6cd
+size 5340675829

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23c2d673ce068efff391a79b76c5ac1c6f55b9e44f28f9be93ff2d8452a5f7b7
+size 595715839

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 115520,
+  "pad_token": "<|im_end|>",
+  "padding_side": "left",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}