Upload folder using huggingface_hub

.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 +1,102 @@
+---
+library_name: mlx
+license: apache-2.0
+license_link: https://huggingface.co/Qwen/Qwen3.5-2B/blob/main/LICENSE
+pipeline_tag: text-generation
+base_model: Qwen/Qwen3.5-2B
+tags:
+- mlx
+- quantized
+- mixed-precision
+- 4bit
+- 8bit
+- optiq
+---
+
+# Qwen3.5-2B-4bit-OptiQ
+
+> Mixed-precision quantized with OptiQ — sensitivity-driven quantization for Apple Silicon
+
+This is a mixed-precision quantized version of [Qwen/Qwen3.5-2B](https://huggingface.co/Qwen/Qwen3.5-2B) in MLX format. Unlike uniform quantization (all layers at the same bit-width), OptiQ measures each layer's sensitivity and assigns optimal per-layer bit-widths, preserving model quality where it matters most.
+
+## How OptiQ Works
+
+OptiQ is an optimizing compiler that converts PyTorch models into hardware-optimized MLX versions using data-driven mixed-precision quantization:
+
+1. **Sensitivity Analysis** — For each layer, OptiQ simulates quantization at each candidate bit-width and measures the KL divergence between the original and quantized output distributions. Layers that distort the distribution more are "sensitive."
+2. **Greedy Knapsack Optimization** — Starting with all layers at the minimum bit-width, OptiQ greedily upgrades the most sensitive layers to higher precision until the target bits-per-weight budget is exhausted.
+3. **Per-Layer Bit Allocation** — The result is a custom quantization config where each layer gets the bit-width that maximizes quality within the size budget. Protected layers (embeddings, final layers) are always assigned high precision.
+
+## Quantization Details
+
+| Property | Value |
+|---|---|
+| Target BPW | 4.5 |
+| Achieved BPW | 4.50 |
+| Candidate bits | 4, 8 |
+| Layers at 4-bit | 116 |
+| Layers at 8-bit | 71 |
+| Total quantized layers | 187 |
+| Group size | 64 |
+| Model size | 1365 MB |
+| Uniform 4-bit size | 1010 MB |
+| Calibration data | WikiText-2 (2 samples, 128 tokens) |
+
+## Benchmark Results
+
+**GSM8K** (200 samples, 3-shot chain-of-thought):
+
+| Model | Candidates | GSM8K Accuracy | Size |
+|---|---|---|---|
+| **OptiQ mixed (4.5 BPW)** | **4, 8** | **48.0%** | **1365 MB** |
+| Uniform 4-bit | — | 48.5% | 1010 MB |
+| OptiQ 3.5 BPW | 3, 4, 8 | 14.0% | 1141 MB |
+| Uniform 3-bit | — | 6.0% | 786 MB |
+| OptiQ 3.0 BPW | 3, 4, 8 | 6.0% | 786 MB |
+| OptiQ 3.0 BPW | 2, 4 | 2.0% | 786 MB |
+| Uniform 2-bit | — | 0.5% | 562 MB |
+
+### Key Findings
+
+At the 2B scale, there is a **sharp quality cliff between 4-bit and 3-bit** — the model drops from ~48% to single digits. This differs from the 0.8B model where the degradation is more gradual and OptiQ's mixed-precision has room to recover quality.
+
+- **At 4-bit (4.5 BPW):** The model is robust. Uniform and mixed-precision perform equivalently — OptiQ provides a safety margin for sensitive layers without measurable cost.
+- **At 3.5 BPW [3,4,8]:** OptiQ achieves 14.0% — 2.3x better than uniform 3-bit (6.0%) — by keeping sensitive layers at 4 or 8-bit while only pushing the least sensitive to 3-bit.
+- **At 3.0 BPW:** The floor bit-width matters enormously. With a 3-bit floor [3,4,8], OptiQ matches uniform 3-bit (6.0%). With a 2-bit floor [2,4], quality collapses to 2.0% — even mixed-precision can't save layers quantized to 2-bit at this scale.
+- **Uniform 2-bit** is essentially random (0.5%).
+
+**Recommendation:** Use this 4.5 BPW model for the best quality-size tradeoff at 2B scale. For smaller models where mixed-precision shows dramatic benefits, see [Qwen3.5-0.8B-4bit-OptiQ](https://huggingface.co/mlx-community/Qwen3.5-0.8B-4bit-OptiQ) where OptiQ more than doubles uniform 4-bit accuracy (27% vs 11.5%).
+
+## Usage
+
+This model works with standard `mlx-lm` — no special code needed:
+
+```python
+from mlx_lm import load, generate
+
+model, tokenizer = load("mlx-community/Qwen3.5-2B-4bit-OptiQ")
+
+prompt = "Explain quantum computing in simple terms:"
+response = generate(model, tokenizer, prompt=prompt, max_tokens=200)
+print(response)
+```
+
+**Requirements:** `mlx-lm >= 0.30.7` (for Qwen3.5 architecture support)
+
+```bash
+pip install mlx-lm>=0.30.7
+```
+
+## Architecture
+
+Qwen3.5 uses a hybrid attention architecture with alternating `linear_attn` and `self_attn` layers. OptiQ's sensitivity analysis identifies which layers are most sensitive to quantization error and assigns them higher precision, while less sensitive layers get 4-bit quantization to minimize model size.
+
+## Article
+
+For more details on the methodology and results, see: [Not All Layers Are Equal: Mixed-Precision Quantization for Weights and KV Cache on Apple Silicon](https://x.com/thin_signal/status/2028412948167942334)
+
+## Credits
+
+- **Quantization method:** OptiQ — optimizing compiler for mixed-precision quantization on Apple Silicon
+- **Base model:** [Qwen/Qwen3.5-2B](https://huggingface.co/Qwen/Qwen3.5-2B) by Qwen Team
+- **Runtime:** [MLX](https://github.com/ml-explore/mlx) by Apple

chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- 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>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- 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>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is true %}
+        {{- '<think>\n' }}
+    {%- else %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

config.json

@@ -0,0 +1,1586 @@
+{
+    "architectures": [
+        "Qwen3_5ForConditionalGeneration"
+    ],
+    "image_token_id": 248056,
+    "model_type": "qwen3_5",
+    "quantization": {
+        "group_size": 64,
+        "bits": 4,
+        "mode": "affine",
+        "language_model.model.embed_tokens": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_qkv": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_z": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_a": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.mlp.gate_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_z": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.mlp.gate_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.q_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.k_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.mlp.gate_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        }
+    },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 4,
+        "mode": "affine",
+        "language_model.model.embed_tokens": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_qkv": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_z": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.in_proj_a": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.mlp.gate_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.0.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_z": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.1.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.mlp.gate_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.2.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.3.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.4.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.5.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.6.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.7.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.8.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.9.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.10.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.11.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.12.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.13.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.14.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.15.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.16.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.17.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.18.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.q_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.k_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.19.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.linear_attn.out_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.20.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.21.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_qkv": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_z": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_b": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.in_proj_a": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.linear_attn.out_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.mlp.gate_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.mlp.down_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.22.mlp.up_proj": {
+            "bits": 4,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.q_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.k_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.v_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.self_attn.o_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.mlp.gate_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.mlp.down_proj": {
+            "bits": 8,
+            "group_size": 64
+        },
+        "language_model.model.layers.23.mlp.up_proj": {
+            "bits": 8,
+            "group_size": 64
+        }
+    },
+    "text_config": {
+        "attention_bias": false,
+        "attention_dropout": 0.0,
+        "attn_output_gate": true,
+        "dtype": "bfloat16",
+        "eos_token_id": 248044,
+        "full_attention_interval": 4,
+        "head_dim": 256,
+        "hidden_act": "silu",
+        "hidden_size": 2048,
+        "initializer_range": 0.02,
+        "intermediate_size": 6144,
+        "layer_types": [
+            "linear_attention",
+            "linear_attention",
+            "linear_attention",
+            "full_attention",
+            "linear_attention",
+            "linear_attention",
+            "linear_attention",
+            "full_attention",
+            "linear_attention",
+            "linear_attention",
+            "linear_attention",
+            "full_attention",
+            "linear_attention",
+            "linear_attention",
+            "linear_attention",
+            "full_attention",
+            "linear_attention",
+            "linear_attention",
+            "linear_attention",
+            "full_attention",
+            "linear_attention",
+            "linear_attention",
+            "linear_attention",
+            "full_attention"
+        ],
+        "linear_conv_kernel_dim": 4,
+        "linear_key_head_dim": 128,
+        "linear_num_key_heads": 16,
+        "linear_num_value_heads": 16,
+        "linear_value_head_dim": 128,
+        "max_position_embeddings": 262144,
+        "mlp_only_layers": [],
+        "model_type": "qwen3_5_text",
+        "mtp_num_hidden_layers": 1,
+        "mtp_use_dedicated_embeddings": false,
+        "num_attention_heads": 8,
+        "num_hidden_layers": 24,
+        "num_key_value_heads": 2,
+        "rms_norm_eps": 1e-06,
+        "tie_word_embeddings": true,
+        "use_cache": true,
+        "vocab_size": 248320,
+        "mamba_ssm_dtype": "float32",
+        "rope_parameters": {
+            "mrope_interleaved": true,
+            "mrope_section": [
+                11,
+                11,
+                10
+            ],
+            "rope_theta": 10000000,
+            "partial_rotary_factor": 0.25,
+            "type": "default"
+        }
+    },
+    "tie_word_embeddings": true,
+    "transformers_version": "4.57.0.dev0",
+    "video_token_id": 248057,
+    "vision_end_token_id": 248054,
+    "vision_start_token_id": 248053
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d47a83c3710cd1183b62bc4df0b0973b2b741a9c8565fac64aee5d69574124ab
+size 1431179061

model.safetensors.index.json

@@ -0,0 +1,702 @@
+{
+    "metadata": {
+        "total_size": 1431089856,
+        "total_parameters": 1881824512
+    },
+    "weight_map": {
+        "language_model.model.embed_tokens.biases": "model.safetensors",
+        "language_model.model.embed_tokens.scales": "model.safetensors",
+        "language_model.model.embed_tokens.weight": "model.safetensors",
+        "language_model.model.layers.0.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.0.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.0.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.0.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.0.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.0.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.0.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.0.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.0.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.0.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.0.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.0.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.1.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.1.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.1.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.1.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.1.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.1.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.1.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.1.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.1.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.1.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.1.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.1.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.10.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.10.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.10.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.10.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.10.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.10.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.10.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.10.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.10.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.10.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.10.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.10.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.11.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.11.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.11.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.11.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.11.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.11.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.11.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.11.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.11.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.11.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.11.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.11.self_attn.k_norm.weight": "model.safetensors",
+        "language_model.model.layers.11.self_attn.k_proj.biases": "model.safetensors",
+        "language_model.model.layers.11.self_attn.k_proj.scales": "model.safetensors",
+        "language_model.model.layers.11.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.model.layers.11.self_attn.o_proj.biases": "model.safetensors",
+        "language_model.model.layers.11.self_attn.o_proj.scales": "model.safetensors",
+        "language_model.model.layers.11.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.model.layers.11.self_attn.q_norm.weight": "model.safetensors",
+        "language_model.model.layers.11.self_attn.q_proj.biases": "model.safetensors",
+        "language_model.model.layers.11.self_attn.q_proj.scales": "model.safetensors",
+        "language_model.model.layers.11.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.model.layers.11.self_attn.v_proj.biases": "model.safetensors",
+        "language_model.model.layers.11.self_attn.v_proj.scales": "model.safetensors",
+        "language_model.model.layers.11.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.model.layers.12.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.12.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.12.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.12.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.12.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.12.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.12.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.12.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.12.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.12.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.12.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.12.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.13.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.13.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.13.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.13.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.13.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.13.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.13.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.13.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.13.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.13.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.13.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.13.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.14.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.14.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.14.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.14.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.14.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.14.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.14.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.14.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.14.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.14.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.14.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.14.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.15.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.15.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.15.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.15.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.15.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.15.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.15.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.15.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.15.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.15.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.15.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.15.self_attn.k_norm.weight": "model.safetensors",
+        "language_model.model.layers.15.self_attn.k_proj.biases": "model.safetensors",
+        "language_model.model.layers.15.self_attn.k_proj.scales": "model.safetensors",
+        "language_model.model.layers.15.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.model.layers.15.self_attn.o_proj.biases": "model.safetensors",
+        "language_model.model.layers.15.self_attn.o_proj.scales": "model.safetensors",
+        "language_model.model.layers.15.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.model.layers.15.self_attn.q_norm.weight": "model.safetensors",
+        "language_model.model.layers.15.self_attn.q_proj.biases": "model.safetensors",
+        "language_model.model.layers.15.self_attn.q_proj.scales": "model.safetensors",
+        "language_model.model.layers.15.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.model.layers.15.self_attn.v_proj.biases": "model.safetensors",
+        "language_model.model.layers.15.self_attn.v_proj.scales": "model.safetensors",
+        "language_model.model.layers.15.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.model.layers.16.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.16.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.16.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.16.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.16.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.16.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.16.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.16.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.16.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.16.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.16.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.16.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.17.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.17.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.17.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.17.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.17.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.17.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.17.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.17.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.17.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.17.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.17.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.17.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.18.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.18.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.18.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.18.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.18.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.18.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.18.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.18.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.18.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.18.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.18.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.18.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.19.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.19.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.19.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.19.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.19.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.19.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.19.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.19.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.19.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.19.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.19.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.19.self_attn.k_norm.weight": "model.safetensors",
+        "language_model.model.layers.19.self_attn.k_proj.biases": "model.safetensors",
+        "language_model.model.layers.19.self_attn.k_proj.scales": "model.safetensors",
+        "language_model.model.layers.19.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.model.layers.19.self_attn.o_proj.biases": "model.safetensors",
+        "language_model.model.layers.19.self_attn.o_proj.scales": "model.safetensors",
+        "language_model.model.layers.19.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.model.layers.19.self_attn.q_norm.weight": "model.safetensors",
+        "language_model.model.layers.19.self_attn.q_proj.biases": "model.safetensors",
+        "language_model.model.layers.19.self_attn.q_proj.scales": "model.safetensors",
+        "language_model.model.layers.19.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.model.layers.19.self_attn.v_proj.biases": "model.safetensors",
+        "language_model.model.layers.19.self_attn.v_proj.scales": "model.safetensors",
+        "language_model.model.layers.19.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.model.layers.2.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.2.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.2.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.2.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.2.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.2.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.2.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.2.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.2.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.2.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.2.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.2.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.20.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.20.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.20.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.20.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.20.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.20.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.20.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.20.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.20.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.20.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.20.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.20.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.21.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.21.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.21.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.21.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.21.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.21.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.21.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.21.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.21.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.21.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.21.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.21.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.22.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.22.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.22.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.22.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.22.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.22.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.22.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.22.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.22.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.22.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.22.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.22.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.23.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.23.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.23.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.23.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.23.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.23.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.23.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.23.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.23.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.23.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.23.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.23.self_attn.k_norm.weight": "model.safetensors",
+        "language_model.model.layers.23.self_attn.k_proj.biases": "model.safetensors",
+        "language_model.model.layers.23.self_attn.k_proj.scales": "model.safetensors",
+        "language_model.model.layers.23.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.model.layers.23.self_attn.o_proj.biases": "model.safetensors",
+        "language_model.model.layers.23.self_attn.o_proj.scales": "model.safetensors",
+        "language_model.model.layers.23.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.model.layers.23.self_attn.q_norm.weight": "model.safetensors",
+        "language_model.model.layers.23.self_attn.q_proj.biases": "model.safetensors",
+        "language_model.model.layers.23.self_attn.q_proj.scales": "model.safetensors",
+        "language_model.model.layers.23.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.model.layers.23.self_attn.v_proj.biases": "model.safetensors",
+        "language_model.model.layers.23.self_attn.v_proj.scales": "model.safetensors",
+        "language_model.model.layers.23.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.model.layers.3.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.3.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.3.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.3.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.3.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.3.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.3.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.3.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.3.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.3.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.3.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.3.self_attn.k_norm.weight": "model.safetensors",
+        "language_model.model.layers.3.self_attn.k_proj.biases": "model.safetensors",
+        "language_model.model.layers.3.self_attn.k_proj.scales": "model.safetensors",
+        "language_model.model.layers.3.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.model.layers.3.self_attn.o_proj.biases": "model.safetensors",
+        "language_model.model.layers.3.self_attn.o_proj.scales": "model.safetensors",
+        "language_model.model.layers.3.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.model.layers.3.self_attn.q_norm.weight": "model.safetensors",
+        "language_model.model.layers.3.self_attn.q_proj.biases": "model.safetensors",
+        "language_model.model.layers.3.self_attn.q_proj.scales": "model.safetensors",
+        "language_model.model.layers.3.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.model.layers.3.self_attn.v_proj.biases": "model.safetensors",
+        "language_model.model.layers.3.self_attn.v_proj.scales": "model.safetensors",
+        "language_model.model.layers.3.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.model.layers.4.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.4.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.4.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.4.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.4.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.4.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.4.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.4.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.4.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.4.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.4.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.4.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.5.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.5.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.5.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.5.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.5.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.5.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.5.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.5.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.5.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.5.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.5.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.5.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.6.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.6.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.6.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.6.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.6.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.6.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.6.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.6.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.6.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.6.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.6.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.6.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.7.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.7.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.7.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.7.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.7.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.7.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.7.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.7.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.7.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.7.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.7.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.7.self_attn.k_norm.weight": "model.safetensors",
+        "language_model.model.layers.7.self_attn.k_proj.biases": "model.safetensors",
+        "language_model.model.layers.7.self_attn.k_proj.scales": "model.safetensors",
+        "language_model.model.layers.7.self_attn.k_proj.weight": "model.safetensors",
+        "language_model.model.layers.7.self_attn.o_proj.biases": "model.safetensors",
+        "language_model.model.layers.7.self_attn.o_proj.scales": "model.safetensors",
+        "language_model.model.layers.7.self_attn.o_proj.weight": "model.safetensors",
+        "language_model.model.layers.7.self_attn.q_norm.weight": "model.safetensors",
+        "language_model.model.layers.7.self_attn.q_proj.biases": "model.safetensors",
+        "language_model.model.layers.7.self_attn.q_proj.scales": "model.safetensors",
+        "language_model.model.layers.7.self_attn.q_proj.weight": "model.safetensors",
+        "language_model.model.layers.7.self_attn.v_proj.biases": "model.safetensors",
+        "language_model.model.layers.7.self_attn.v_proj.scales": "model.safetensors",
+        "language_model.model.layers.7.self_attn.v_proj.weight": "model.safetensors",
+        "language_model.model.layers.8.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.8.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.8.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.8.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.8.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.8.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.8.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.8.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.8.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.8.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.8.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.8.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.9.input_layernorm.weight": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.A_log": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.conv1d.weight": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.dt_bias": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_a.biases": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_a.scales": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_a.weight": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_b.biases": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_b.scales": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_b.weight": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_qkv.biases": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_qkv.scales": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_qkv.weight": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_z.biases": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_z.scales": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.in_proj_z.weight": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.norm.weight": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.out_proj.biases": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.out_proj.scales": "model.safetensors",
+        "language_model.model.layers.9.linear_attn.out_proj.weight": "model.safetensors",
+        "language_model.model.layers.9.mlp.down_proj.biases": "model.safetensors",
+        "language_model.model.layers.9.mlp.down_proj.scales": "model.safetensors",
+        "language_model.model.layers.9.mlp.down_proj.weight": "model.safetensors",
+        "language_model.model.layers.9.mlp.gate_proj.biases": "model.safetensors",
+        "language_model.model.layers.9.mlp.gate_proj.scales": "model.safetensors",
+        "language_model.model.layers.9.mlp.gate_proj.weight": "model.safetensors",
+        "language_model.model.layers.9.mlp.up_proj.biases": "model.safetensors",
+        "language_model.model.layers.9.mlp.up_proj.scales": "model.safetensors",
+        "language_model.model.layers.9.mlp.up_proj.weight": "model.safetensors",
+        "language_model.model.layers.9.post_attention_layernorm.weight": "model.safetensors",
+        "language_model.model.norm.weight": "model.safetensors"
+    }
+}

optiq_metadata.json

@@ -0,0 +1,758 @@
+{
+  "method": "optiq_mixed_precision",
+  "target_bpw": 4.5,
+  "achieved_bpw": 4.49961680484916,
+  "n_high_bits": 71,
+  "n_low_bits": 116,
+  "threshold": 0.0,
+  "per_layer": {
+    "model.layers.0.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.0.linear_attn.in_proj_qkv": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.0.linear_attn.in_proj_z": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.0.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.0.linear_attn.in_proj_a": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.0.mlp.gate_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.0.mlp.up_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.0.mlp.down_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.1.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.1.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.1.linear_attn.in_proj_z": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.1.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.1.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.1.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.1.mlp.up_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.1.mlp.down_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.2.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.2.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.2.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.2.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.2.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.2.mlp.gate_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.2.mlp.up_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.2.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.3.self_attn.q_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.3.self_attn.k_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.3.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.3.self_attn.o_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.3.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.3.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.3.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.4.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.4.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.4.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.4.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.4.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.4.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.4.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.4.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.5.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.5.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.5.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.5.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.5.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.5.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.5.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.5.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.6.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.6.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.6.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.6.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.6.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.6.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.6.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.6.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.7.self_attn.q_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.7.self_attn.k_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.7.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.7.self_attn.o_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.7.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.7.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.7.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.8.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.8.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.8.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.8.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.8.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.8.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.8.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.8.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.9.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.9.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.9.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.9.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.9.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.9.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.9.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.9.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.10.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.10.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.10.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.10.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.10.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.10.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.10.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.10.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.11.self_attn.q_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.11.self_attn.k_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.11.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.11.self_attn.o_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.11.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.11.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.11.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.12.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.12.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.12.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.12.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.12.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.12.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.12.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.12.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.13.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.13.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.13.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.13.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.13.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.13.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.13.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.13.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.14.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.14.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.14.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.14.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.14.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.14.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.14.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.14.mlp.down_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.15.self_attn.q_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.15.self_attn.k_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.15.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.15.self_attn.o_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.15.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.15.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.15.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.16.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.16.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.16.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.16.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.16.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.16.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.16.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.16.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.17.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.17.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.17.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.17.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.17.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.17.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.17.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.17.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.18.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.18.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.18.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.18.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.18.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.18.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.18.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.18.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.19.self_attn.q_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.19.self_attn.k_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.19.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.19.self_attn.o_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.19.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.19.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.19.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.20.linear_attn.out_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.20.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.20.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.20.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.20.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.20.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.20.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.20.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.21.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.21.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.21.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.21.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.21.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.21.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.21.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.21.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.22.linear_attn.out_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.22.linear_attn.in_proj_qkv": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.22.linear_attn.in_proj_z": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.22.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.22.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.22.mlp.gate_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.22.mlp.up_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.22.mlp.down_proj": {
+      "bits": 4,
+      "group_size": 64
+    },
+    "model.layers.23.self_attn.q_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.23.self_attn.k_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.23.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.23.self_attn.o_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.23.mlp.gate_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.23.mlp.up_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "model.layers.23.mlp.down_proj": {
+      "bits": 8,
+      "group_size": 64
+    },
+    "lm_head": {
+      "bits": 4,
+      "group_size": 64
+    }
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
+size 19989343

tokenizer_config.json

@@ -0,0 +1,32 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": true,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "split_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "tool_parser_type": "qwen3_coder",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}