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 +1,56 @@
+---
+library_name: transformers
+license: other
+license_name: nvidia-open-model-license
+license_link: https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/
+pipeline_tag: text-generation
+language:
+- en
+- es
+- fr
+- de
+- ja
+- it
+tags:
+- nvidia
+- pytorch
+- mlx
+datasets:
+- nvidia/Nemotron-Pretraining-Code-v1
+- nvidia/Nemotron-CC-v2
+- nvidia/Nemotron-Pretraining-SFT-v1
+- nvidia/Nemotron-CC-Math-v1
+- nvidia/Nemotron-Pretraining-Code-v2
+- nvidia/Nemotron-Pretraining-Specialized-v1
+- nvidia/Nemotron-CC-v2.1
+- nvidia/Nemotron-CC-Code-v1
+- nvidia/Nemotron-Pretraining-Dataset-sample
+- nvidia/Nemotron-Competitive-Programming-v1
+- nvidia/Nemotron-Math-v2
+- nvidia/Nemotron-Agentic-v1
+- nvidia/Nemotron-Math-Proofs-v1
+- nvidia/Nemotron-Instruction-Following-Chat-v1
+- nvidia/Nemotron-Science-v1
+- nvidia/Nemotron-3-Nano-RL-Training-Blend
+track_downloads: true
+base_model: nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
+---
+## 💫 Community Model> NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 by nvidia
+
+_👾 [LM Studio](https://lmstudio.ai) Community models highlights program. Highlighting new & noteworthy models by the community. Join the conversation on [Discord](https://discord.gg/aPQfnNkxGC)_.
+
+**Model creator**: [nvidia](https://huggingface.co/nvidia)<br>
+**Original model**: [NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16)<br>
+**MLX quantization**: provided by [LM Studio team](https://x.com/lmstudio) using [mlx_lm](https://github.com/ml-explore/mlx-lm)<br>
+
+## Technical Details
+
+6-bit quantized version of NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 using MLX, optimized for Apple Silicon.
+
+## Special thanks
+
+🙏 Special thanks to the [Apple Machine Learning Research](https://github.com/ml-explore) team for creating [MLX](https://github.com/ml-explore/mlx).
+
+## Disclaimers
+
+LM Studio is not the creator, originator, or owner of any Model featured in the Community Model Program. Each Community Model is created and provided by third parties. LM Studio does not endorse, support, represent or guarantee the completeness, truthfulness, accuracy, or reliability of any Community Model. You understand that Community Models can produce content that might be offensive, harmful, inaccurate or otherwise inappropriate, or deceptive. Each Community Model is the sole responsibility of the person or entity who originated such Model. LM Studio may not monitor or control the Community Models and cannot, and does not, take responsibility for any such Model. LM Studio disclaims all warranties or guarantees about the accuracy, reliability or benefits of the Community Models. LM Studio further disclaims any warranty that the Community Model will meet your requirements, be secure, uninterrupted or available at any time or location, or error-free, viruses-free, or that any errors will be corrected, or otherwise. You will be solely responsible for any damage resulting from your use of or access to the Community Models, your downloading of any Community Model, or use of any other Community Model provided by or through LM Studio.

chat_template.jinja

@@ -0,0 +1,204 @@
+{% macro render_extra_keys(json_dict, handled_keys) %}
+    {%- if json_dict is mapping %}
+        {%- for json_key in json_dict if json_key not in handled_keys %}
+            {%- if json_dict[json_key] is mapping or (json_dict[json_key] is sequence and json_dict[json_key] is not string) %}
+                {{- '\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | tojson | safe) ~ '</' ~ json_key ~ '>' }}
+            {%- else %}
+                {{-'\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | string) ~ '</' ~ json_key ~ '>' }}
+            {%- endif %}
+        {%- endfor %}
+    {%- endif %}
+{% endmacro %}
+{%- set enable_thinking = enable_thinking if enable_thinking is defined else True %}
+{%- set truncate_history_thinking = truncate_history_thinking if truncate_history_thinking is defined else True %}
+
+{%- set ns = namespace(last_user_idx = -1) %}
+{%- set loop_messages = messages %}
+{%- for m in loop_messages %}
+  {%- if m["role"] == "user" %}
+    {%- set ns.last_user_idx = loop.index0 %}
+  {%- endif %}
+{%- endfor %}
+
+{%- if messages[0]["role"] == "system" %}
+    {%- set system_message = messages[0]["content"] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set system_message = "" %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = [] %}
+{%- endif %}
+{# Recompute last_user_idx relative to loop_messages after handling system #}
+{%- set ns = namespace(last_user_idx = -1) %}
+{%- for m in loop_messages %}
+  {%- if m["role"] == "user" %}
+    {%- set ns.last_user_idx = loop.index0 %}
+  {%- endif %}
+{%- endfor %}
+{%- if system_message is defined %}
+    {{- "<|im_start|>system\n" + system_message }}
+{%- else %}
+    {%- if tools is iterable and tools | length > 0 %}
+        {{- "<|im_start|>system\n" }}
+    {%- endif %}
+{%- endif %}
+{%- if tools is iterable and tools | length > 0 %}
+    {%- if system_message is defined and system_message | length > 0 %}
+        {{- "\n\n" }}
+    {%- endif %}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n" }}
+    {{- "<tools>" }}
+    {%- for tool in tools %}
+        {%- if tool.function is defined %}
+            {%- set tool = tool.function %}
+        {%- endif %}
+        {{- "\n<function>\n<name>" ~ tool.name ~ "</name>" }}
+        {%- if tool.description is defined %}
+            {{- '\n<description>' ~ (tool.description | trim) ~ '</description>' }}
+        {%- endif %}
+        {{- '\n<parameters>' }}
+        {%- if tool.parameters is defined and tool.parameters is mapping and tool.parameters.properties is defined and tool.parameters.properties is mapping %}
+            {%- for param_name, param_fields in tool.parameters.properties|items %}
+                {{- '\n<parameter>' }}
+                {{- '\n<name>' ~ param_name ~ '</name>' }}
+                {%- if param_fields.type is defined %}
+                    {{- '\n<type>' ~ (param_fields.type | string) ~ '</type>' }}
+                {%- endif %}
+                {%- if param_fields.description is defined %}
+                    {{- '\n<description>' ~ (param_fields.description | trim) ~ '</description>' }}
+                {%- endif %}
+                {%- if param_fields.enum is defined %}
+                    {{- '\n<enum>' ~ (param_fields.enum | tojson | safe) ~ '</enum>' }}
+                {%- endif %}
+                {%- set handled_keys = ['name', 'type', 'description', 'enum'] %}
+                {{- render_extra_keys(param_fields, handled_keys) }}
+                {{- '\n</parameter>' }}
+            {%- endfor %}
+        {%- endif %}
+        {% set handled_keys = ['type', 'properties', 'required'] %}
+        {{- render_extra_keys(tool.parameters, handled_keys) }}
+        {%- if tool.parameters is defined and tool.parameters.required is defined %}
+            {{- '\n<required>' ~ (tool.parameters.required | tojson | safe) ~ '</required>' }}
+        {%- endif %}
+        {{- '\n</parameters>' }}
+        {%- set handled_keys = ['type', 'name', 'description', 'parameters'] %}
+        {{- render_extra_keys(tool, handled_keys) }}
+        {{- '\n</function>' }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+{%- endif %}
+
+
+{%- if system_message is defined %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if tools is iterable and tools | length > 0 %}
+        {{- '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+
+{%- for message in loop_messages %}
+    {%- if message.role == "assistant" %}
+        {# Add reasoning content in to content field for unified processing below. #}
+        {%- if message.reasoning_content is defined and message.reasoning_content is string and message.reasoning_content | trim | length > 0 %}
+            {%- set content = "<think>\n" ~ message.reasoning_content ~ "\n</think>\n" ~ (message.content | default('', true)) %}
+        {%- else %}
+            {%- set content = message.content | default('', true) %}
+            {%- if content is string -%}
+                {# Allow downstream logic to to take care of broken thought, only handle coherent reasoning here. #}
+                {%- if '<think>' not in content and '</think>' not in content -%}
+                    {%- set content = "<think></think>" ~ content -%}
+                {%- endif -%}
+            {%- else -%}
+                {%- set content = content -%}
+            {%- endif -%}
+        {%- endif %}
+        {%- if message.tool_calls is defined and message.tool_calls is iterable and message.tool_calls | length > 0 %}
+            {# Assistant message has tool calls. #}
+            {{- '<|im_start|>assistant\n' }}
+                {%- set include_content = not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
+                {%- if content is string and content | trim | length > 0 %}
+                    {%- if include_content %}
+                        {{- (content | trim) ~ '\n' -}}
+                    {%- else %}
+                        {%- set c = (content | string) %}
+                        {%- if '</think>' in c %}
+                            {# Keep only content after the last closing think. Also generation prompt causes this. #}
+                            {%- set c = c.split('</think>')[-1] %}
+                        {%- elif '<think>' in c %}
+                            {# If <think> was opened but never closed, drop the trailing think segment #}
+                            {%- set c = c.split('<think>')[0] %}
+                        {%- endif %}
+                        {%- set c = "<think></think>" ~ c | trim %}
+                        {%- if c | length > 0 %}
+                            {{- c ~ '\n' -}}
+                        {%- endif %}
+                    {%- endif %}
+                {%- else %}
+                    {{- "<think></think>" -}}
+                {%- endif %}
+                {%- for tool_call in message.tool_calls %}
+                    {%- if tool_call.function is defined %}
+                        {%- set tool_call = tool_call.function %}
+                    {%- endif %}
+                    {{- '<tool_call>\n<function=' ~ tool_call.name ~ '>\n' -}}
+                        {%- if tool_call.arguments is defined %}
+                            {%- for args_name, args_value in tool_call.arguments|items %}
+                                {{- '<parameter=' ~ args_name ~ '>\n' -}}
+                                    {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                                {{- args_value ~ '\n</parameter>\n' -}}
+                            {%- endfor %}
+                        {%- endif %}
+                    {{- '</function>\n</tool_call>\n' -}}
+                {%- endfor %}
+                {{- '<|im_end|>\n' }}
+        {%- else %}
+            {# Assistant message doesn't have tool calls. #}
+            {%- if not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}
+                {{- '<|im_start|>assistant\n' ~ (content | default('', true) | string | trim) ~ '<|im_end|>\n' }}
+            {%- else %}
+                {%- set c = (content | default('', true) | string) %}
+                {%- if '<think>' in c and '</think>' in c %}
+                    {%- set c = "<think></think>" ~ c.split('</think>')[-1] %}
+                {%- endif %}
+                {%- set c = c | trim %}
+                {%- if c | length > 0 %}
+                    {{- '<|im_start|>assistant\n' ~ c ~ '<|im_end|>\n' }}
+                {%- else %}
+                    {{- '<|im_start|>assistant\n<|im_end|>\n' }}
+                {%- endif %}
+            {%- endif %}
+        {%- endif %}
+    {%- elif message.role == "user" or message.role == "system" %}
+        {{- '<|im_start|>' + message.role + '\n' }}
+        {%- set content = message.content | string %}
+        {{- content }}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user\n' }}
+        {%- endif %}
+        {{- '<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>\n' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endfor %}
+
+{%- if add_generation_prompt %}
+    {%- if enable_thinking %}
+        {{- '<|im_start|>assistant\n<think>\n' }}
+    {%- else %}
+        {{- '<|im_start|>assistant\n<think></think>' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,80 @@
+{
+    "architectures": [
+        "NemotronHForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "auto_map": {
+        "AutoConfig": "configuration_nemotron_h.NemotronHConfig",
+        "AutoModel": "modeling_nemotron_h.NemotronHForCausalLM",
+        "AutoModelForCausalLM": "modeling_nemotron_h.NemotronHForCausalLM"
+    },
+    "bos_token_id": 1,
+    "chunk_size": 128,
+    "conv_kernel": 4,
+    "eos_token_id": 2,
+    "expand": 2,
+    "head_dim": 128,
+    "hidden_dropout": 0.0,
+    "hidden_size": 2688,
+    "hybrid_override_pattern": "MEMEM*EMEMEM*EMEMEM*EMEMEM*EMEMEM*EMEMEMEM*EMEMEMEME",
+    "initializer_range": 0.02,
+    "intermediate_size": 1856,
+    "layer_norm_epsilon": 1e-05,
+    "mamba_head_dim": 64,
+    "mamba_hidden_act": "silu",
+    "mamba_num_heads": 64,
+    "mamba_proj_bias": false,
+    "mamba_ssm_cache_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "mlp_bias": false,
+    "mlp_hidden_act": "relu2",
+    "model_type": "nemotron_h",
+    "moe_intermediate_size": 1856,
+    "moe_shared_expert_intermediate_size": 3712,
+    "n_group": 1,
+    "n_groups": 8,
+    "n_routed_experts": 128,
+    "n_shared_experts": 1,
+    "norm_eps": 1e-05,
+    "norm_topk_prob": true,
+    "num_attention_heads": 32,
+    "num_experts_per_tok": 6,
+    "num_hidden_layers": 52,
+    "num_key_value_heads": 2,
+    "num_logits_to_keep": 1,
+    "pad_token_id": 0,
+    "partial_rotary_factor": 1.0,
+    "quantization": {
+        "group_size": 64,
+        "bits": 6,
+        "mode": "affine"
+    },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 6,
+        "mode": "affine"
+    },
+    "rescale_prenorm_residual": true,
+    "residual_in_fp32": false,
+    "rope_theta": 10000,
+    "routed_scaling_factor": 2.5,
+    "sliding_window": null,
+    "ssm_state_size": 128,
+    "tie_word_embeddings": false,
+    "time_step_floor": 0.0001,
+    "time_step_limit": [
+        0.0,
+        Infinity
+    ],
+    "time_step_max": 0.1,
+    "time_step_min": 0.001,
+    "topk_group": 1,
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.55.4",
+    "use_bias": false,
+    "use_cache": true,
+    "use_conv_bias": true,
+    "use_mamba_kernels": true,
+    "vocab_size": 131072
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "do_sample": true,
+  "bos_token_id": 1,
+  "eos_token_id": [2, 11],
+  "pad_token_id": 0,
+  "temperature": 1.0,
+  "top_p": 1.0,
+  "transformers_version": "4.55.4"
+}

model-00001-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96bc1bc806daa4a932ef633b7a1d084a60dcde2d984dcf976ccf0dd8458dc523
+size 5200192765

model-00002-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15bdaadbeefc7e5504b95d8c7359a23bde85aaff75f74a22198c360cdb6a769d
+size 4949851395

model-00003-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7ce3c1009440654f2136386a52bae801ea0fed2e6bbae48a1714bf372b6d8a7
+size 5361336506

model-00004-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a8e24a45b2d7562366f629a202e1e827630db10a3f3fbe07526761e24e8fd63
+size 4989763011

model-00005-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efe8f5675018907d70cbf1fdbbb7b479bc024642d8f13ca7acbf25269ff4fd7e
+size 5166553607

model.safetensors.index.json

@@ -0,0 +1,737 @@
+{
+    "metadata": {
+        "total_size": 25667611392,
+        "total_parameters": 31577935872
+    },
+    "weight_map": {
+        "backbone.embeddings.biases": "model-00001-of-00005.safetensors",
+        "backbone.embeddings.scales": "model-00001-of-00005.safetensors",
+        "backbone.embeddings.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.A_log": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.D": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.conv1d.bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.conv1d.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.dt_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.in_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.in_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.in_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.out_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.out_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.mixer.out_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.0.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.gate.e_score_correction_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.gate.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.shared_experts.down_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.shared_experts.down_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.shared_experts.down_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.shared_experts.up_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.shared_experts.up_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.shared_experts.up_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.switch_mlp.fc1.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.switch_mlp.fc1.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.switch_mlp.fc1.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.switch_mlp.fc2.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.switch_mlp.fc2.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.mixer.switch_mlp.fc2.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.1.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.10.mixer.gate.e_score_correction_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.gate.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.shared_experts.down_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.shared_experts.down_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.shared_experts.down_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.shared_experts.up_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.shared_experts.up_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.shared_experts.up_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.switch_mlp.fc1.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.10.mixer.switch_mlp.fc1.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.10.mixer.switch_mlp.fc1.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.10.mixer.switch_mlp.fc2.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.switch_mlp.fc2.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.mixer.switch_mlp.fc2.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.10.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.11.mixer.A_log": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.D": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.conv1d.bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.conv1d.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.dt_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.in_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.in_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.in_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.out_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.out_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.mixer.out_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.11.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.k_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.k_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.k_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.o_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.o_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.o_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.q_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.q_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.q_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.v_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.v_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.mixer.v_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.12.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.gate.e_score_correction_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.gate.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.shared_experts.down_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.shared_experts.down_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.shared_experts.down_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.shared_experts.up_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.shared_experts.up_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.shared_experts.up_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.switch_mlp.fc1.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.switch_mlp.fc1.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.switch_mlp.fc1.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.switch_mlp.fc2.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.switch_mlp.fc2.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.mixer.switch_mlp.fc2.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.13.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.A_log": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.D": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.conv1d.bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.conv1d.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.dt_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.in_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.in_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.in_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.out_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.out_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.mixer.out_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.14.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.gate.e_score_correction_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.gate.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.shared_experts.down_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.shared_experts.down_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.shared_experts.down_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.shared_experts.up_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.shared_experts.up_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.shared_experts.up_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.switch_mlp.fc1.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.switch_mlp.fc1.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.switch_mlp.fc1.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.switch_mlp.fc2.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.switch_mlp.fc2.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.mixer.switch_mlp.fc2.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.15.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.A_log": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.D": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.conv1d.bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.conv1d.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.dt_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.in_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.in_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.in_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.out_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.out_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.mixer.out_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.16.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.gate.e_score_correction_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.gate.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.shared_experts.down_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.shared_experts.down_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.shared_experts.down_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.shared_experts.up_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.shared_experts.up_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.shared_experts.up_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.switch_mlp.fc1.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.switch_mlp.fc1.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.switch_mlp.fc1.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.switch_mlp.fc2.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.switch_mlp.fc2.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.mixer.switch_mlp.fc2.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.17.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.A_log": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.D": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.conv1d.bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.conv1d.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.dt_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.in_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.in_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.in_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.out_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.out_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.mixer.out_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.18.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.k_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.k_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.k_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.o_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.o_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.o_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.q_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.q_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.q_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.v_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.v_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.mixer.v_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.19.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.2.mixer.A_log": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.D": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.conv1d.bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.conv1d.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.dt_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.in_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.in_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.in_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.out_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.out_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.mixer.out_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.2.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.20.mixer.gate.e_score_correction_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.gate.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.shared_experts.down_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.shared_experts.down_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.shared_experts.down_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.shared_experts.up_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.shared_experts.up_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.shared_experts.up_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.switch_mlp.fc1.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.switch_mlp.fc1.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.switch_mlp.fc1.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.switch_mlp.fc2.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.switch_mlp.fc2.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.mixer.switch_mlp.fc2.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.20.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.A_log": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.D": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.conv1d.bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.conv1d.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.dt_bias": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.in_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.in_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.in_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.out_proj.biases": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.out_proj.scales": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.mixer.out_proj.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.21.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.22.mixer.gate.e_score_correction_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.gate.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.shared_experts.down_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.shared_experts.down_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.shared_experts.down_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.shared_experts.up_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.shared_experts.up_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.shared_experts.up_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.switch_mlp.fc1.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.switch_mlp.fc1.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.switch_mlp.fc1.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.switch_mlp.fc2.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.switch_mlp.fc2.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.mixer.switch_mlp.fc2.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.22.norm.weight": "model-00002-of-00005.safetensors",
+        "backbone.layers.23.mixer.A_log": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.D": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.conv1d.bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.conv1d.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.dt_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.in_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.in_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.in_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.out_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.out_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.mixer.out_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.23.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.gate.e_score_correction_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.gate.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.shared_experts.down_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.shared_experts.down_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.shared_experts.down_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.shared_experts.up_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.shared_experts.up_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.shared_experts.up_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.switch_mlp.fc1.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.switch_mlp.fc1.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.switch_mlp.fc1.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.switch_mlp.fc2.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.switch_mlp.fc2.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.mixer.switch_mlp.fc2.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.24.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.A_log": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.D": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.conv1d.bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.conv1d.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.dt_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.in_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.in_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.in_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.out_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.out_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.mixer.out_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.25.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.k_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.k_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.k_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.o_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.o_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.o_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.q_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.q_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.q_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.v_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.v_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.mixer.v_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.26.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.gate.e_score_correction_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.gate.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.shared_experts.down_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.shared_experts.down_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.shared_experts.down_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.shared_experts.up_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.shared_experts.up_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.shared_experts.up_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.switch_mlp.fc1.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.switch_mlp.fc1.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.switch_mlp.fc1.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.switch_mlp.fc2.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.switch_mlp.fc2.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.mixer.switch_mlp.fc2.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.27.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.A_log": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.D": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.conv1d.bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.conv1d.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.dt_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.in_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.in_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.in_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.out_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.out_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.mixer.out_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.28.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.gate.e_score_correction_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.gate.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.shared_experts.down_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.shared_experts.down_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.shared_experts.down_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.shared_experts.up_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.shared_experts.up_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.shared_experts.up_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.switch_mlp.fc1.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.switch_mlp.fc1.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.switch_mlp.fc1.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.switch_mlp.fc2.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.switch_mlp.fc2.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.mixer.switch_mlp.fc2.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.29.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.3.mixer.gate.e_score_correction_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.gate.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.shared_experts.down_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.shared_experts.down_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.shared_experts.down_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.shared_experts.up_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.shared_experts.up_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.shared_experts.up_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.switch_mlp.fc1.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.switch_mlp.fc1.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.switch_mlp.fc1.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.switch_mlp.fc2.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.switch_mlp.fc2.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.mixer.switch_mlp.fc2.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.3.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.30.mixer.A_log": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.D": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.conv1d.bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.conv1d.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.dt_bias": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.in_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.in_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.in_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.out_proj.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.out_proj.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.mixer.out_proj.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.30.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.31.mixer.gate.e_score_correction_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.gate.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.shared_experts.down_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.shared_experts.down_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.shared_experts.down_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.shared_experts.up_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.shared_experts.up_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.shared_experts.up_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.switch_mlp.fc1.biases": "model-00003-of-00005.safetensors",
+        "backbone.layers.31.mixer.switch_mlp.fc1.scales": "model-00003-of-00005.safetensors",
+        "backbone.layers.31.mixer.switch_mlp.fc1.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.31.mixer.switch_mlp.fc2.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.switch_mlp.fc2.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.31.mixer.switch_mlp.fc2.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.31.norm.weight": "model-00003-of-00005.safetensors",
+        "backbone.layers.32.mixer.A_log": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.D": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.conv1d.bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.conv1d.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.dt_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.in_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.in_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.in_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.out_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.out_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.mixer.out_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.32.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.k_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.k_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.k_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.o_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.o_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.o_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.q_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.q_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.q_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.v_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.v_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.mixer.v_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.33.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.gate.e_score_correction_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.gate.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.shared_experts.down_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.shared_experts.down_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.shared_experts.down_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.shared_experts.up_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.shared_experts.up_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.shared_experts.up_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.switch_mlp.fc1.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.switch_mlp.fc1.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.switch_mlp.fc1.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.switch_mlp.fc2.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.switch_mlp.fc2.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.mixer.switch_mlp.fc2.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.34.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.A_log": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.D": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.conv1d.bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.conv1d.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.dt_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.in_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.in_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.in_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.out_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.out_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.mixer.out_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.35.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.gate.e_score_correction_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.gate.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.shared_experts.down_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.shared_experts.down_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.shared_experts.down_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.shared_experts.up_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.shared_experts.up_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.shared_experts.up_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.switch_mlp.fc1.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.switch_mlp.fc1.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.switch_mlp.fc1.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.switch_mlp.fc2.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.switch_mlp.fc2.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.mixer.switch_mlp.fc2.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.36.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.A_log": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.D": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.conv1d.bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.conv1d.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.dt_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.in_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.in_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.in_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.out_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.out_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.mixer.out_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.37.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.gate.e_score_correction_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.gate.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.shared_experts.down_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.shared_experts.down_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.shared_experts.down_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.shared_experts.up_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.shared_experts.up_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.shared_experts.up_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.switch_mlp.fc1.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.switch_mlp.fc1.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.switch_mlp.fc1.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.switch_mlp.fc2.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.switch_mlp.fc2.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.mixer.switch_mlp.fc2.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.38.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.A_log": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.D": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.conv1d.bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.conv1d.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.dt_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.in_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.in_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.in_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.out_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.out_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.mixer.out_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.39.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.4.mixer.A_log": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.D": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.conv1d.bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.conv1d.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.dt_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.in_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.in_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.in_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.out_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.out_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.mixer.out_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.4.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.40.mixer.gate.e_score_correction_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.gate.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.shared_experts.down_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.shared_experts.down_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.shared_experts.down_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.shared_experts.up_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.shared_experts.up_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.shared_experts.up_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.switch_mlp.fc1.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.switch_mlp.fc1.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.switch_mlp.fc1.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.switch_mlp.fc2.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.switch_mlp.fc2.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.mixer.switch_mlp.fc2.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.40.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.A_log": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.D": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.conv1d.bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.conv1d.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.dt_bias": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.in_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.in_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.in_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.out_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.out_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.mixer.out_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.41.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.k_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.k_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.k_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.o_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.o_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.o_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.q_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.q_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.q_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.v_proj.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.v_proj.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.mixer.v_proj.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.42.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.43.mixer.gate.e_score_correction_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.gate.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.shared_experts.down_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.shared_experts.down_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.shared_experts.down_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.shared_experts.up_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.shared_experts.up_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.shared_experts.up_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.switch_mlp.fc1.biases": "model-00004-of-00005.safetensors",
+        "backbone.layers.43.mixer.switch_mlp.fc1.scales": "model-00004-of-00005.safetensors",
+        "backbone.layers.43.mixer.switch_mlp.fc1.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.43.mixer.switch_mlp.fc2.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.switch_mlp.fc2.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.mixer.switch_mlp.fc2.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.43.norm.weight": "model-00004-of-00005.safetensors",
+        "backbone.layers.44.mixer.A_log": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.D": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.conv1d.bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.conv1d.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.dt_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.in_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.in_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.in_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.out_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.out_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.mixer.out_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.44.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.gate.e_score_correction_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.gate.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.shared_experts.down_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.shared_experts.down_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.shared_experts.down_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.shared_experts.up_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.shared_experts.up_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.shared_experts.up_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.switch_mlp.fc1.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.switch_mlp.fc1.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.switch_mlp.fc1.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.switch_mlp.fc2.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.switch_mlp.fc2.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.mixer.switch_mlp.fc2.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.45.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.A_log": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.D": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.conv1d.bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.conv1d.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.dt_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.in_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.in_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.in_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.out_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.out_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.mixer.out_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.46.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.gate.e_score_correction_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.gate.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.shared_experts.down_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.shared_experts.down_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.shared_experts.down_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.shared_experts.up_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.shared_experts.up_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.shared_experts.up_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.switch_mlp.fc1.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.switch_mlp.fc1.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.switch_mlp.fc1.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.switch_mlp.fc2.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.switch_mlp.fc2.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.mixer.switch_mlp.fc2.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.47.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.A_log": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.D": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.conv1d.bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.conv1d.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.dt_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.in_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.in_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.in_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.out_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.out_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.mixer.out_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.48.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.gate.e_score_correction_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.gate.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.shared_experts.down_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.shared_experts.down_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.shared_experts.down_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.shared_experts.up_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.shared_experts.up_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.shared_experts.up_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.switch_mlp.fc1.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.switch_mlp.fc1.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.switch_mlp.fc1.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.switch_mlp.fc2.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.switch_mlp.fc2.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.mixer.switch_mlp.fc2.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.49.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.5.mixer.k_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.k_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.k_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.o_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.o_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.o_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.q_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.q_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.q_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.v_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.v_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.mixer.v_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.5.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.50.mixer.A_log": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.D": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.conv1d.bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.conv1d.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.dt_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.in_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.in_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.in_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.out_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.out_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.mixer.out_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.50.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.gate.e_score_correction_bias": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.gate.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.shared_experts.down_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.shared_experts.down_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.shared_experts.down_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.shared_experts.up_proj.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.shared_experts.up_proj.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.shared_experts.up_proj.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.switch_mlp.fc1.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.switch_mlp.fc1.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.switch_mlp.fc1.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.switch_mlp.fc2.biases": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.switch_mlp.fc2.scales": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.mixer.switch_mlp.fc2.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.51.norm.weight": "model-00005-of-00005.safetensors",
+        "backbone.layers.6.mixer.gate.e_score_correction_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.gate.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.shared_experts.down_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.shared_experts.down_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.shared_experts.down_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.shared_experts.up_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.shared_experts.up_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.shared_experts.up_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.switch_mlp.fc1.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.switch_mlp.fc1.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.switch_mlp.fc1.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.switch_mlp.fc2.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.switch_mlp.fc2.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.mixer.switch_mlp.fc2.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.6.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.A_log": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.D": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.conv1d.bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.conv1d.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.dt_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.in_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.in_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.in_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.out_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.out_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.mixer.out_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.7.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.gate.e_score_correction_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.gate.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.shared_experts.down_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.shared_experts.down_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.shared_experts.down_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.shared_experts.up_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.shared_experts.up_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.shared_experts.up_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.switch_mlp.fc1.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.switch_mlp.fc1.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.switch_mlp.fc1.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.switch_mlp.fc2.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.switch_mlp.fc2.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.mixer.switch_mlp.fc2.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.8.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.A_log": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.D": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.conv1d.bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.conv1d.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.dt_bias": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.in_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.in_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.in_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.out_proj.biases": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.out_proj.scales": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.mixer.out_proj.weight": "model-00001-of-00005.safetensors",
+        "backbone.layers.9.norm.weight": "model-00001-of-00005.safetensors",
+        "backbone.norm_f.weight": "model-00005-of-00005.safetensors",
+        "lm_head.biases": "model-00005-of-00005.safetensors",
+        "lm_head.scales": "model-00005-of-00005.safetensors",
+        "lm_head.weight": "model-00005-of-00005.safetensors"
+    }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0722467d08447bcaf49b3bbc04af5b7055fc9e14965075b870c88e1d088418f3
+size 17077478

tokenizer_config.json

@@ -0,0 +1,18 @@
+{
+  "add_prefix_space": false,
+  "additional_special_tokens": null,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "is_local": true,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {},
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<unk>",
+  "chat_template": "{% macro render_extra_keys(json_dict, handled_keys) %}\n    {%- if json_dict is mapping %}\n        {%- for json_key in json_dict if json_key not in handled_keys %}\n            {%- if json_dict[json_key] is mapping or (json_dict[json_key] is sequence and json_dict[json_key] is not string) %}\n                {{- '\\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | tojson | safe) ~ '</' ~ json_key ~ '>' }}\n            {%- else %}\n                {{-'\\n<' ~ json_key ~ '>' ~ (json_dict[json_key] | string) ~ '</' ~ json_key ~ '>' }}\n            {%- endif %}\n        {%- endfor %}\n    {%- endif %}\n{% endmacro %}\n{%- set enable_thinking = enable_thinking if enable_thinking is defined else True %}\n{%- set truncate_history_thinking = truncate_history_thinking if truncate_history_thinking is defined else True %}\n\n{%- set ns = namespace(last_user_idx = -1) %}\n{%- set loop_messages = messages %}\n{%- for m in loop_messages %}\n  {%- if m[\"role\"] == \"user\" %}\n    {%- set ns.last_user_idx = loop.index0 %}\n  {%- endif %}\n{%- endfor %}\n\n{%- if messages[0][\"role\"] == \"system\" %}\n    {%- set system_message = messages[0][\"content\"] %}\n    {%- set loop_messages = messages[1:] %}\n{%- else %}\n    {%- set system_message = \"\" %}\n    {%- set loop_messages = messages %}\n{%- endif %}\n{%- if not tools is defined %}\n    {%- set tools = [] %}\n{%- endif %}\n{# Recompute last_user_idx relative to loop_messages after handling system #}\n{%- set ns = namespace(last_user_idx = -1) %}\n{%- for m in loop_messages %}\n  {%- if m[\"role\"] == \"user\" %}\n    {%- set ns.last_user_idx = loop.index0 %}\n  {%- endif %}\n{%- endfor %}\n{%- if system_message is defined %}\n    {{- \"<|im_start|>system\\n\" + system_message }}\n{%- else %}\n    {%- if tools is iterable and tools | length > 0 %}\n        {{- \"<|im_start|>system\\n\" }}\n    {%- endif %}\n{%- endif %}\n{%- if tools is iterable and tools | length > 0 %}\n    {%- if system_message is defined and system_message | length > 0 %}\n        {{- \"\\n\\n\" }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou have access to the following functions:\\n\\n\" }}\n    {{- \"<tools>\" }}\n    {%- for tool in tools %}\n        {%- if tool.function is defined %}\n            {%- set tool = tool.function %}\n        {%- endif %}\n        {{- \"\\n<function>\\n<name>\" ~ tool.name ~ \"</name>\" }}\n        {%- if tool.description is defined %}\n            {{- '\\n<description>' ~ (tool.description | trim) ~ '</description>' }}\n        {%- endif %}\n        {{- '\\n<parameters>' }}\n        {%- if tool.parameters is defined and tool.parameters is mapping and tool.parameters.properties is defined and tool.parameters.properties is mapping %}\n            {%- for param_name, param_fields in tool.parameters.properties|items %}\n                {{- '\\n<parameter>' }}\n                {{- '\\n<name>' ~ param_name ~ '</name>' }}\n                {%- if param_fields.type is defined %}\n                    {{- '\\n<type>' ~ (param_fields.type | string) ~ '</type>' }}\n                {%- endif %}\n                {%- if param_fields.description is defined %}\n                    {{- '\\n<description>' ~ (param_fields.description | trim) ~ '</description>' }}\n                {%- endif %}\n                {%- if param_fields.enum is defined %}\n                    {{- '\\n<enum>' ~ (param_fields.enum | tojson | safe) ~ '</enum>' }}\n                {%- endif %}\n                {%- set handled_keys = ['name', 'type', 'description', 'enum'] %}\n                {{- render_extra_keys(param_fields, handled_keys) }}\n                {{- '\\n</parameter>' }}\n            {%- endfor %}\n        {%- endif %}\n        {% set handled_keys = ['type', 'properties', 'required'] %}\n        {{- render_extra_keys(tool.parameters, handled_keys) }}\n        {%- if tool.parameters is defined and tool.parameters.required is defined %}\n            {{- '\\n<required>' ~ (tool.parameters.required | tojson | safe) ~ '</required>' }}\n        {%- endif %}\n        {{- '\\n</parameters>' }}\n        {%- set handled_keys = ['type', 'name', 'description', 'parameters'] %}\n        {{- render_extra_keys(tool, handled_keys) }}\n        {{- '\\n</function>' }}\n    {%- endfor %}\n    {{- \"\\n</tools>\" }}\n\n    {{- '\\n\\nIf you choose to call a function ONLY reply in the following format with NO suffix:\\n\\n<tool_call>\\n<function=example_function_name>\\n<parameter=example_parameter_1>\\nvalue_1\\n</parameter>\\n<parameter=example_parameter_2>\\nThis is the value for the second parameter\\nthat can span\\nmultiple lines\\n</parameter>\\n</function>\\n</tool_call>\\n\\n<IMPORTANT>\\nReminder:\\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\\n- Required parameters MUST be specified\\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\\n</IMPORTANT>' }}\n{%- endif %}\n\n\n{%- if system_message is defined %}\n    {{- '<|im_end|>\\n' }}\n{%- else %}\n    {%- if tools is iterable and tools | length > 0 %}\n        {{- '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n\n{%- for message in loop_messages %}\n    {%- if message.role == \"assistant\" %}\n        {# Add reasoning content in to content field for unified processing below. #}\n        {%- if message.reasoning_content is defined and message.reasoning_content is string and message.reasoning_content | trim | length > 0 %}\n            {%- set content = \"<think>\\n\" ~ message.reasoning_content ~ \"\\n</think>\\n\" ~ (message.content | default('', true)) %}\n        {%- else %}\n            {%- set content = message.content | default('', true) %}\n            {%- if content is string -%}\n                {# Allow downstream logic to to take care of broken thought, only handle coherent reasoning here. #}\n                {%- if '<think>' not in content and '</think>' not in content -%}\n                    {%- set content = \"<think></think>\" ~ content -%}\n                {%- endif -%}\n            {%- else -%}\n                {%- set content = content -%}\n            {%- endif -%}\n        {%- endif %}\n        {%- if message.tool_calls is defined and message.tool_calls is iterable and message.tool_calls | length > 0 %}\n            {# Assistant message has tool calls. #}\n            {{- '<|im_start|>assistant\\n' }}\n                {%- set include_content = not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}\n                {%- if content is string and content | trim | length > 0 %}\n                    {%- if include_content %}\n                        {{- (content | trim) ~ '\\n' -}}\n                    {%- else %}\n                        {%- set c = (content | string) %}\n                        {%- if '</think>' in c %}\n                            {# Keep only content after the last closing think. Also generation prompt causes this. #}\n                            {%- set c = c.split('</think>')[-1] %}\n                        {%- elif '<think>' in c %}\n                            {# If <think> was opened but never closed, drop the trailing think segment #}\n                            {%- set c = c.split('<think>')[0] %}\n                        {%- endif %}\n                        {%- set c = \"<think></think>\" ~ c | trim %}\n                        {%- if c | length > 0 %}\n                            {{- c ~ '\\n' -}}\n                        {%- endif %}\n                    {%- endif %}\n                {%- else %}\n                    {{- \"<think></think>\" -}}\n                {%- endif %}\n                {%- for tool_call in message.tool_calls %}\n                    {%- if tool_call.function is defined %}\n                        {%- set tool_call = tool_call.function %}\n                    {%- endif %}\n                    {{- '<tool_call>\\n<function=' ~ tool_call.name ~ '>\\n' -}}\n                        {%- if tool_call.arguments is defined %}\n                            {%- for args_name, args_value in tool_call.arguments|items %}\n                                {{- '<parameter=' ~ args_name ~ '>\\n' -}}\n                                    {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}\n                                {{- args_value ~ '\\n</parameter>\\n' -}}\n                            {%- endfor %}\n                        {%- endif %}\n                    {{- '</function>\\n</tool_call>\\n' -}}\n                {%- endfor %}\n                {{- '<|im_end|>\\n' }}\n        {%- else %}\n            {# Assistant message doesn't have tool calls. #}\n            {%- if not (truncate_history_thinking and loop.index0 < ns.last_user_idx) %}\n                {{- '<|im_start|>assistant\\n' ~ (content | default('', true) | string | trim) ~ '<|im_end|>\\n' }}\n            {%- else %}\n                {%- set c = (content | default('', true) | string) %}\n                {%- if '<think>' in c and '</think>' in c %}\n                    {%- set c = \"<think></think>\" ~ c.split('</think>')[-1] %}\n                {%- endif %}\n                {%- set c = c | trim %}\n                {%- if c | length > 0 %}\n                    {{- '<|im_start|>assistant\\n' ~ c ~ '<|im_end|>\\n' }}\n                {%- else %}\n                    {{- '<|im_start|>assistant\\n<|im_end|>\\n' }}\n                {%- endif %}\n            {%- endif %}\n        {%- endif %}\n    {%- elif message.role == \"user\" or message.role == \"system\" %}\n        {{- '<|im_start|>' + message.role + '\\n' }}\n        {%- set content = message.content | string %}\n        {{- content }}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.previtem and loop.previtem.role != \"tool\" %}\n            {{- '<|im_start|>user\\n' }}\n        {%- endif %}\n        {{- '<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>\\n' }}\n        {%- if not loop.last and loop.nextitem.role != \"tool\" %}\n            {{- '<|im_end|>\\n' }}\n        {%- elif loop.last %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- else %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endfor %}\n\n{%- if add_generation_prompt %}\n    {%- if enable_thinking %}\n        {{- '<|im_start|>assistant\\n<think>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>assistant\\n<think></think>' }}\n    {%- endif %}\n{%- endif %}\n"
+}