初始化项目，由ModelHub XC社区提供模型

Model: dicta-il/DictaLM-3.0-24B-Thinking-W4A16
Source: Original Platform

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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 +1,86 @@
+---
+license: apache-2.0
+pipeline_tag: text-generation
+language:
+  - en
+  - he
+tags:
+- pretrained
+inference:
+  parameters:
+    temperature: 0.6
+---
+
+[<img src="https://i.ibb.co/5Lbwyr1/dicta-logo.jpg" width="300px"/>](https://dicta.org.il)
+
+# Dicta-LM 3.0: Advancing The Frontier of Hebrew Sovereign LLMs 
+
+Dicta-LM 3.0 is a powerful open-weight collection of LLMs, trained on extensive corpora of Hebrew and English texts. The models are available for download and for unlimited use. The models set a new SOTA for their weight-class for Hebrew, both as base models and chat models.
+
+This is our flagship model, a 24-billion-parameter *reasoning* model, originally initialized from [Mistral-Small-3.1-24B-Base-2503](https://huggingface.co/mistralai/Mistral-Small-3.1-24B-Base-2503).
+
+This model is a reasoning chat model, which means that before responding to any given message from the user, the model first thinks out the right way to respond in a designated thinking block. 
+
+This version of the model is quantized to 4-bits (with 16-bit activations), allowing for inference with significantly less memory although with slightly weaker performance. This version of the model can fit on a single 24GB GPU. 
+
+<br/>
+
+&#128640; Try it out in full precision here: [chat.dicta.org.il](https://chat.dicta.org.il)
+
+<br/>
+
+For full details of this model please read our [release blog post](https://dicta.org.il/dicta-lm-3) or the [technical report](https://www.dicta.org.il/publications/DictaLM_3_0___Techincal_Report.pdf).
+
+You can view and access the full collection of base/instruct unquantized/quantized versions of `DictaLM 3.0` [here](https://huggingface.co/collections/dicta-il/dictalm-30-collection).
+
+## Instruction format
+
+In order to leverage instruction fine-tuning, your prompt should be rendered using the chat template specified for this model. Most libraries deal with this automatically, so you can just let them do it. 
+
+## Usage
+
+### vLLM
+
+```bash
+vllm serve dicta-il/DictaLM-3.0-24B-Thinking-W4A16 --enable-auto-tool-choice --tool-call-parser hermes --reasoning_parser deepseek_r1
+```
+
+> If you run out of memory on a 24GB GPU, decrease the context window and enforce eager: `--max-model-len 8192 --enforce-eager`
+
+And then you can access it via the openai library:
+
+```python
+from openai import OpenAI
+
+client = OpenAI(
+    base_url="http://localhost:8000/v1",
+    api_key="sk-no-key-required"
+)
+
+response = client.chat.completions.create(
+    model="dicta-il/DictaLM-3.0-24B-Thinking-W4A16",
+    messages=[
+        {"role": "user", "content": "Hello, how are you?"}
+    ],
+)
+
+print(response.choices[0].message.content)
+```
+
+> The reasoning traces should be available in the response structure in the designated fild. 
+
+The model supports tool-calling, enabling integration with external tools and APIs. For example how to use the tool calling, see the [vLLM documentation](https://docs.vllm.ai/en/stable/features/tool_calling/#tool-calling).
+
+## Citation
+
+If you use this model, please cite:
+
+```bibtex
+@article{Shmidman2025DictaLM3,
+  title={{Dicta-LM 3.0: Advancing The Frontier of Hebrew Sovereign LLMs}},
+  author={Shaltiel Shmidman and Avi Shmidman and Amir DN Cohen and Moshe Koppel},
+  year={2025},
+  publisher={{DICTA / Jerusalem, Israel}},
+  note={https://www.dicta.org.il/publications/DictaLM_3_0___Techincal_Report.pdf}
+}
+```

chat_template.jinja

@@ -0,0 +1,92 @@
+{#  ───── header (system message) ─────  #}
+{{- "<|im_start|>system\n" -}}
+{#  ───── get the custom instructions + optional thinking override ─────  #}
+{%- if messages[0].role == "system" -%}
+	{%- set custom_instructions = messages[0].content.rstrip() -%}
+{%- endif -%}
+{#  ───── set the system prompt ─────  #}
+{%- if custom_instructions -%}
+	{{- custom_instructions -}}
+{%- else -%}
+	{{- "You are a helpful AI assistant named Dicta-LM 3.0, Trained by Dicta, the Israel Center for Text Analysis. Your role is to provide accurate, helpful, and well-structured responses to user questions and requests.\nProvide clear, logical, and precise answers that thoroughly address what the user is asking for. Structure your responses in a way that is easy to understand and follow." -}}
+{%- 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" and message.content is string and not (message.content.startswith("<tool_response>") and message.content.endswith("</tool_response>")) -%}
+		{%- set ns.multi_step_tool = false -%}
+		{%- set ns.last_query_index = index -%}
+	{%- endif -%}
+{%- endfor -%}
+{%- if tools -%}
+	{{- "\n\n## Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" -}}
+	{%- for tool in tools -%}
+		{{- "\n" -}}
+		{{- tool | tojson -}}
+	{%- endfor -%}
+	{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call>" -}}
+{%- endif -%}
+{{- "<|im_end|>" -}}
+{%- for message in messages -%}
+	{%- if message.content is string -%}
+		{%- set content = message.content -%}
+	{%- else -%}
+		{%- set content = "" -%}
+	{%- endif -%}
+	{%- if message.role == "user" or message.role == "system" and not loop.first -%}
+		{{- "\n<|im_start|>" + message.role + "\n" + content + "<|im_end|>" -}}
+	{%- elif message.role == "assistant" -%}
+		{% generation %}
+		{%- set reasoning_content = "" -%}
+		{%- if message.reasoning_content is string -%}
+			{%- set reasoning_content = message.reasoning_content -%}
+		{%- elif "</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 -%}
+		{%- if loop.index0 > ns.last_query_index -%}
+			{%- if loop.last or not loop.last and reasoning_content -%}
+				{{- "\n<|im_start|>" + message.role + "\n<think>" + reasoning_content.strip("\n") + "</think>" + content.lstrip("\n") -}}
+			{%- else -%}
+				{{- "\n<|im_start|>" + message.role + "\n" + content -}}
+			{%- endif -%}
+		{%- else -%}
+			{{- "\n<|im_start|>" + message.role + "\n" + content -}}
+		{%- endif -%}
+		{%- if message.tool_calls -%}
+			{%- for tool_call in message.tool_calls -%}
+				{%- if loop.first and content or not loop.first -%}
+					{{- "\n" -}}
+				{%- endif -%}
+				{%- if tool_call.function -%}
+					{%- set tool_call = tool_call.function -%}
+				{%- endif -%}
+				{{- "<tool_call>\n{\"name\": \"" -}}
+				{{- tool_call.name -}}
+				{{- "\", \"arguments\": " -}}
+				{%- if tool_call.arguments is string -%}
+					{{- tool_call.arguments -}}
+				{%- else -%}
+					{{- tool_call.arguments | tojson -}}
+				{%- endif -%}
+				{{- "}\n</tool_call>" -}}
+			{%- endfor -%}
+		{%- endif -%}
+		{{- "<|im_end|>" -}}
+		{% endgeneration %}
+	{%- elif message.role == "tool" -%}
+		{%- if loop.first or messages[loop.index0 - 1].role != "tool" -%}
+			{{- "\n<|im_start|>user" -}}
+		{%- endif -%}
+		{{- "\n<tool_response>\n" -}}
+		{{- content -}}
+		{{- "\n</tool_response>" -}}
+		{%- if loop.last or messages[loop.index0 + 1].role != "tool" -%}
+			{{- "<|im_end|>" -}}
+		{%- endif -%}
+	{%- endif -%}
+{%- endfor -%}
+{{- "\n" -}}
+{%- if add_generation_prompt -%}
+	{{- "<|im_start|>assistant\n<think>" -}}
+{%- endif -%}

config.json

@@ -0,0 +1,61 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 32768,
+  "max_position_embeddings": 65280,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 8,
+  "quantization_config": {
+    "config_groups": {
+      "group_0": {
+        "format": "pack-quantized",
+        "input_activations": null,
+        "output_activations": null,
+        "targets": [
+          "Linear"
+        ],
+        "weights": {
+          "actorder": "static",
+          "block_structure": null,
+          "dynamic": false,
+          "group_size": 128,
+          "num_bits": 4,
+          "observer": "minmax",
+          "observer_kwargs": {},
+          "strategy": "group",
+          "symmetric": true,
+          "type": "int"
+        }
+      }
+    },
+    "format": "pack-quantized",
+    "global_compression_ratio": null,
+    "ignore": [
+      "lm_head"
+    ],
+    "kv_cache_scheme": null,
+    "quant_method": "compressed-tensors",
+    "quantization_status": "compressed",
+    "sparsity_config": {},
+    "transform_config": {},
+    "version": "0.12.2"
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.56.2",
+  "use_cache": true,
+  "vocab_size": 131072
+}

generation_config.json

@@ -0,0 +1,15 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": [
+    2,
+    21
+  ],
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "frequency_penalty": 0.1,
+  "transformers_version": "4.56.2"
+}

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca04ee07020a269e75167ef18f4eba07d6c4472b1c7e3f48890565a067373bea
+size 4981180656

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7b64f97254a5d0709875f59adef00504b32ca667b2e8c49c910bdcce23fc5ed
+size 4985413200

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d84fb9e63f0c0b38a5fb297402ead851a656a01cfcdb5fbab2fa504c1b626b77
+size 4180946336

model.safetensors.index.json

@@ -0,0 +1,931 @@
+{
+  "metadata": {
+    "total_parameters": 4294989360,
+    "total_size": 14147434880
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight_packed": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight_scale": "model-00002-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight_shape": "model-00002-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.32.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.32.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.33.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.33.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.34.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.34.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.35.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.35.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.36.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.36.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.37.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.37.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.38.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.38.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.down_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.down_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.down_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.gate_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.gate_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.gate_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.up_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.up_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.39.mlp.up_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.k_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.k_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.k_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.o_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.o_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.o_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.q_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.q_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.q_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.v_proj.weight_packed": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.v_proj.weight_scale": "model-00003-of-00003.safetensors",
+    "model.layers.39.self_attn.v_proj.weight_shape": "model-00003-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight_packed": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight_scale": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight_shape": "model-00001-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

recipe.yaml

@@ -0,0 +1,20 @@
+default_stage:
+  default_modifiers:
+    SmoothQuantModifier:
+      smoothing_strength: 0.8
+      mappings:
+      - !!python/tuple
+        - ['re:.*q_proj', 're:.*k_proj', 're:.*v_proj']
+        - re:.*input_layernorm
+      - !!python/tuple
+        - ['re:.*gate_proj', 're:.*up_proj']
+        - re:.*post_attention_layernorm
+      ignore: []
+    GPTQModifier:
+      targets: [Linear]
+      ignore: [lm_head]
+      scheme: W4A16
+      block_size: 128
+      dampening_frac: 0.01
+      actorder: static
+      offload_hessians: false

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d54d32258647e5a4e66e7a211bd9a8794674666b4ebb049ccf851161a9e489cd
+size 17078033