moss-image-mlx 0.1.0 release

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/en_sign.jpg filter=lfs diff=lfs merge=lfs -text
+tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

NOTICE

@@ -0,0 +1,14 @@
+moss-image-mlx
+Copyright 2026 Felix Brener. Licensed under the Apache License, Version 2.0.
+
+This project is an independent MLX implementation of MOSS-Image. It is not affiliated with or endorsed by OpenMOSS.
+
+It reproduces the inference behaviour of software and models released by others:
+- MOSS-Image (model weights and pipeline_moss_image.py), Copyright OpenMOSS, Apache License 2.0.
+  https://github.com/OpenMOSS/MOSS-Image
+- Qwen3.5-9B (the text encoder weights MOSS-Image ships), Copyright Alibaba Cloud (Qwen team), Apache License 2.0.
+- FLUX.2 VAE with the FLUX.2 small decoder (the autoencoder MOSS-Image ships), Copyright Black Forest Labs,
+  Apache License 2.0. https://huggingface.co/black-forest-labs/FLUX.2-small-decoder
+- mlx-lm (used as a dependency for the Qwen3.5 backbone), Copyright Apple Inc., MIT License.
+
+torch_randn.py and the strip-decoding VAE are adapted from felix-ab/flux3-mlx (Apache License 2.0).

README.md

@@ -0,0 +1,118 @@
+---
+license: apache-2.0
+base_model: OpenMOSS-Team/MOSS-Image-SFT
+base_model_relation: quantized
+library_name: mlx
+pipeline_tag: text-to-image
+language:
+- en
+- zh
+tags:
+- mlx
+- apple-silicon
+- moss-image
+- text-to-image
+- text-rendering
+- layout
+- quantized
+- 8-bit
+---
+
+# MOSS-Image-SFT for Apple Silicon (MLX, 8bit)
+
+An MLX build of [OpenMOSS-Team/MOSS-Image-SFT](https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT), OpenMOSS's 5B text-to-image DiT for long bilingual text rendering and JSON layout control, for Macs with Apple Silicon. Runs with [moss-image-mlx](https://github.com/felix-ab/moss-image-mlx), a port verified against the official PyTorch pipeline.
+
+- **9.31 GB peak memory** for a 1024² image with a short prompt (about +1 GB with a 650-token prompt), against a 27 GB download and a 16.6 GB peak for the original bf16 weights.
+- **text and layout identical to bf16 on every eval prompt** (14-image evaluation below).
+- **Official seeds reproduce official images.**
+- **Download:** 14.1 GB. Transformer 8-bit; text encoder (Qwen3.5-9B language path) 8-bit; VAE decoder fp32.
+
+<p align="center"><img src="assets/en_sign.jpg" width="32%"> <img src="assets/zh_poster.jpg" width="32%"> <img src="assets/layout_jazz.jpg" width="32%"></p>
+<p align="center"><i>Renders from this tier at 1024², 50 steps, CFG 3, seed 42. More in assets/.</i></p>
+
+## Quickstart
+
+```bash
+pip install moss-image-mlx
+moss-mlx "A cozy bakery at dusk. A wooden sign reads \"THE GOLDEN CRUST BAKERY\"." --model VisualInference/MOSS-Image-SFT-MLX-8bit -o bakery.png
+```
+
+```python
+from moss_image_mlx import MOSSImagePipeline, Request, save
+from huggingface_hub import snapshot_download
+
+pipe = MOSSImagePipeline(snapshot_download("VisualInference/MOSS-Image-SFT-MLX-8bit"))
+req = Request(prompt="A neon sign that reads \"OPEN LATE\" above a ramen bar", width=1024, height=1024, seed=7)
+save(pipe(req), req, "ramen.png")
+```
+
+Layout prompts (JSON), batch rendering, the prompt cache and every option are documented in the [GitHub README](https://github.com/felix-ab/moss-image-mlx).
+
+## Which tier?
+
+| Tier | Download | Peak at 1024² | Pick it if |
+|---|---|---|---|
+| [8bit](https://huggingface.co/VisualInference/MOSS-Image-SFT-MLX-8bit) | 14.1 GB | 9.3 GB | you have 24 GB or more; reference quality |
+| [compact](https://huggingface.co/VisualInference/MOSS-Image-SFT-MLX-compact) | 12.1 GB | 7.4 GB | you have 16 GB |
+| original bf16 (`--model OpenMOSS-Team/MOSS-Image-SFT`) | 27.2 GB | 16.6 GB | you have 32 GB or more; fastest (~11.5 min per image), no quantization |
+
+Measured on a Mac mini M4 Pro (48 GB), MLX 0.32.4, one 1024² image at the official defaults (50 steps, CFG 3, shift 5, block 22 skipped in the negative branch). Time: **12.4 min** per image. The model is compute-bound, so quantization saves memory, not time.
+
+## Quality evidence
+
+The same 14 prompts were rendered with the bf16 weights and with this tier, then scored:
+- **OCR:** Apple Vision, order-free character recall against the requested text.
+- **Layout:** IoU between each requested text box and where the text actually appears.
+
+| Prompt (1024², seed 42) | bf16 reference | This tier |
+|---|---|---|
+| Bakery sign + chalkboard (EN): OCR recall | 0.98 | 0.98 |
+| Book cover, 3 lines (EN): OCR recall | 1.00 | 1.00 |
+| Typed 4-line paragraph (EN): OCR recall | 1.00 | 1.00 |
+| 3-step infographic (EN): OCR recall | 0.99 | 0.99 |
+| Handwritten fridge note (ZH): OCR recall | 1.00 | 1.00 |
+| Tea-house calligraphy sign (ZH) ¹: OCR recall | 0.30 | 0.30 |
+| Event poster, 3 lines (ZH): OCR recall | 1.00 | 1.00 |
+| Printed notice, 4 lines (ZH): OCR recall | 1.00 | 1.00 |
+| Layout JSON: EXPLORE poster: box IoU per text | 0.96 | 0.98 |
+| Layout JSON: jazz poster, 3 text boxes: box IoU per text | 0.88, 0.93, 0.66 | 0.88, 0.93, 0.61 |
+
+**Summary:**
+- OCR mean 0.908 (bf16: 0.908).
+- Layout IoU 0.850 (bf16: 0.858).
+- Mean PSNR vs the bf16 render 30.8 dB. With this model, bf16 rounding alone moves an image about 23 dB.
+
+¹ Apple's OCR cannot read the vertical calligraphy in this prompt for *any* variant, bf16 included; the render itself is correct.
+
+The remaining misses were checked by eye and are OCR misreads of correct renders (for example 顺 read as 順). More: [how the port was verified](https://github.com/felix-ab/moss-image-mlx#how-it-was-verified).
+
+**What is quantized:**
+- **Transformer:** the 42 transformer blocks' linear layers use MLX affine 8-bit weights (group size 64). The patch embedder, timestep MLP, shared modulation, head modulation, final projection and text projection stay bf16 (~1% of weights).
+- **Text encoder:** linear layers and embeddings use 8-bit, group size 64. Its RMSNorm weights are kept in fp32 to match the reference.
+- **VAE:** the decoder only, fp32.
+
+## Limitations
+
+- Dense, very small text can garble, as in the original model.
+- The SFT checkpoint shows faint vertical banding on some skies and walls (upstream behaviour).
+- ~12.4 min per 1024² image on an M4 Pro at the official 50 steps. Few-step support is planned.
+
+## 中文说明
+
+本仓库是 OpenMOSS **MOSS-Image-SFT** 的 Apple Silicon（MLX）量化版本，配合 [moss-image-mlx](https://github.com/felix-ab/moss-image-mlx) 使用。
+
+- **显存与下载：** 生成一张 1024² 图像的峰值内存约 **9.31 GB**，下载 14.1 GB。
+- **正确性：** 已与官方 PyTorch 流程逐项对齐（采样日程逐位一致，fp32 端到端 69.7 dB）。
+- **种子复现：** 使用官方种子可复现官方示例图。
+- **文字渲染：** 在 14 张图的评测中，中英文文字渲染与 bf16 原版完全一致。
+
+安装：`pip install moss-image-mlx`，然后运行 `moss-mlx "提示词" --model VisualInference/MOSS-Image-SFT-MLX-8bit`。
+
+## License and attribution
+
+Apache-2.0, the same as the original weights. See [NOTICE](NOTICE).
+- **MOSS-Image:** © OpenMOSS.
+- **Text encoder:** Qwen3.5-9B (Apache-2.0).
+- **VAE:** FLUX.2 with the FLUX.2 small decoder by Black Forest Labs (Apache-2.0).
+
+This is an independent conversion, not affiliated with or endorsed by OpenMOSS.

config.json

@@ -0,0 +1,16 @@
+{
+  "model_type": "moss_image",
+  "library_name": "moss-image-mlx",
+  "format": 1,
+  "tier": "8bit",
+  "source_model": "OpenMOSS-Team/MOSS-Image-SFT",
+  "transformer_quantization": "affine8g64",
+  "text_encoder_quantization": "affine8g64",
+  "components": [
+    "transformer",
+    "text_encoder",
+    "tokenizer",
+    "vae",
+    "scheduler"
+  ]
+}

model_index.json

@@ -0,0 +1,27 @@
+{
+  "_class_name": "MOSSImagePipeline",
+  "_diffusers_version": "0.38.0",
+  "scheduler": [
+    "pipeline_moss_image",
+    "MOSSImageScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "Qwen3_5Model"
+  ],
+  "tokenizer": [
+    "transformers",
+    "Qwen2Tokenizer"
+  ],
+  "transformer": [
+    "pipeline_moss_image",
+    "MOSSImageTransformer"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLFlux2"
+  ],
+  "moss_image_mlx": {
+    "format": 1
+  }
+}

scheduler/scheduler_config.json

@@ -0,0 +1,19 @@
+{
+  "_class_name": "MOSSImageScheduler",
+  "_diffusers_version": "0.38.0",
+  "base_image_seq_len": 256,
+  "base_shift": 0.5,
+  "invert_sigmas": false,
+  "max_image_seq_len": 4096,
+  "max_shift": 1.15,
+  "num_train_timesteps": 1000,
+  "shift": 5.0,
+  "shift_terminal": null,
+  "stochastic_sampling": false,
+  "time_shift_type": "exponential",
+  "training_shift": 1.0,
+  "use_beta_sigmas": false,
+  "use_dynamic_shifting": false,
+  "use_exponential_sigmas": false,
+  "use_karras_sigmas": false
+}

text_encoder/config.json

@@ -0,0 +1,99 @@
+{
+  "architectures": [
+    "Qwen3_5TextModel"
+  ],
+  "dtype": "bfloat16",
+  "image_token_id": 248056,
+  "model_type": "qwen3_5",
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_output_gate": true,
+    "bos_token_id": null,
+    "dtype": "bfloat16",
+    "eos_token_id": 248044,
+    "full_attention_interval": 4,
+    "head_dim": 256,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 12288,
+    "layer_types": [
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention"
+    ],
+    "linear_conv_kernel_dim": 4,
+    "linear_key_head_dim": 128,
+    "linear_num_key_heads": 16,
+    "linear_num_value_heads": 32,
+    "linear_value_head_dim": 128,
+    "mamba_ssm_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "mlp_only_layers": [],
+    "model_type": "qwen3_5_text",
+    "mtp_num_hidden_layers": 1,
+    "mtp_use_dedicated_embeddings": false,
+    "num_attention_heads": 16,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 4,
+    "pad_token_id": null,
+    "partial_rotary_factor": 0.25,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        11,
+        11,
+        10
+      ],
+      "partial_rotary_factor": 0.25,
+      "rope_theta": 10000000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vocab_size": 248320
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.12.1",
+  "video_token_id": 248057,
+  "vision_end_token_id": 248054,
+  "vision_start_token_id": 248053,
+  "moss_image_mlx": {
+    "format": 1,
+    "quantization": {
+      "bits": 8,
+      "group_size": 64
+    }
+  }
+}

text_encoder/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a471ed08fefe259aab5e3f7c24602759fe344cedfece4a10682066cbf43f97c9
+size 4249064728

text_encoder/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e3cbd59f2743435c9f0eaf00858102f6998f2c6d02bc38bef9078cfb49ef711
+size 4185299477

text_encoder/model.safetensors.index.json

@@ -0,0 +1,931 @@
+{
+ "metadata": {
+  "total_size": 8434263040
+ },
+ "weight_map": {
+  "embed_tokens.weight": "model-00001-of-00002.safetensors",
+  "embed_tokens.scales": "model-00001-of-00002.safetensors",
+  "embed_tokens.biases": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.0.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.0.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.1.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.1.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.2.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.2.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.3.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.4.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.4.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.5.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.5.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.6.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.6.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.7.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.8.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.8.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.9.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.9.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.10.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.10.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.q_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.k_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.v_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.o_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.11.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.12.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.12.mlp.up_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.conv1d.weight": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_qkv.weight": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_qkv.scales": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_qkv.biases": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_z.weight": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_z.scales": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_z.biases": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_b.weight": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_b.scales": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_b.biases": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_a.weight": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_a.scales": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.in_proj_a.biases": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.dt_bias": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.A_log": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.norm.weight": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.out_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.13.linear_attn.out_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+  "layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.13.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.13.mlp.gate_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+  "layers.13.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+  "layers.13.mlp.down_proj.biases": "model-00001-of-00002.safetensors",
+  "layers.13.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.13.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.13.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.14.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.14.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.15.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.16.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.16.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.17.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.17.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.18.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.18.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.19.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.20.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.20.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.21.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.21.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.22.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.22.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.23.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.24.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.24.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.25.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.25.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.26.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.26.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.27.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.28.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.28.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.29.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.29.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.conv1d.weight": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_qkv.weight": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_qkv.scales": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_qkv.biases": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_z.weight": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_z.scales": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_z.biases": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_b.weight": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_b.scales": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_b.biases": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_a.weight": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_a.scales": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.in_proj_a.biases": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.dt_bias": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.A_log": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.norm.weight": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.out_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.30.linear_attn.out_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.30.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.q_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.k_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.v_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.o_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.gate_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.down_proj.biases": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+  "layers.31.mlp.up_proj.biases": "model-00002-of-00002.safetensors",
+  "norm.weight": "model-00002-of-00002.safetensors"
+ }
+}

tokenizer/chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
+size 19989325

tokenizer/tokenizer_config.json

@@ -0,0 +1,32 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": true,
+  "local_files_only": true,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

transformer/config.json

@@ -0,0 +1,19 @@
+{
+  "_class_name": "MOSSImageTransformer",
+  "_diffusers_version": "0.38.0",
+  "dim": 3072,
+  "ffn_dim": 8192,
+  "freq_dim": 256,
+  "latent_channels": 128,
+  "num_heads": 24,
+  "num_layers": 42,
+  "rope_base": 10000.0,
+  "text_dim": 4096,
+  "moss_image_mlx": {
+    "format": 1,
+    "quantization": {
+      "bits": 8,
+      "group_size": 64
+    }
+  }
+}

transformer/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f010b8a8f18dfb8217f943c566bb1cba6a4f37bd3fe7a75da4fae479f619d10f
+size 4291488455

transformer/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade726e8d81139c634dd0b01054a4198127624bd4ebabaa936b5ce8802d59174
+size 1201822635

transformer/model.safetensors.index.json

@@ -0,0 +1,993 @@
+{
+ "metadata": {
+  "total_size": 5493211136
+ },
+ "weight_map": {
+  "video_embedder.weight": "model-00001-of-00002.safetensors",
+  "video_embedder.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.proj.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.proj.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.norm1.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.norm2.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.q_norm.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.k_norm.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.qkv.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.qkv.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.qkv.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.qkv.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.proj.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.proj.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.proj.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.proj.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_gate.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_gate.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_gate.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_gate.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_up.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_up.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_up.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_up.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.0.ffn_down.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.norm1.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.norm2.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.q_norm.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.k_norm.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.qkv.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.qkv.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.qkv.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.qkv.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.proj.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.proj.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.proj.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.proj.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_gate.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_gate.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_gate.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_gate.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_up.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_up.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_up.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_up.bias": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "text_embedder.blocks.1.ffn_down.bias": "model-00001-of-00002.safetensors",
+  "time_embedding.0.weight": "model-00001-of-00002.safetensors",
+  "time_embedding.0.bias": "model-00001-of-00002.safetensors",
+  "time_embedding.1.weight": "model-00001-of-00002.safetensors",
+  "time_embedding.1.bias": "model-00001-of-00002.safetensors",
+  "shared_block_modulation.weight": "model-00001-of-00002.safetensors",
+  "shared_block_modulation.bias": "model-00001-of-00002.safetensors",
+  "layers.0.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.0.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.0.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.0.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.0.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.0.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.0.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.0.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.0.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.0.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.0.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.0.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.0.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.0.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.0.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.0.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.0.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.0.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.0.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.0.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.0.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.0.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.1.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.1.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.1.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.1.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.1.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.1.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.1.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.1.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.1.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.1.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.1.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.1.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.1.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.1.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.1.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.1.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.1.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.1.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.1.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.1.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.1.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.1.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.2.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.2.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.2.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.2.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.2.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.2.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.2.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.2.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.2.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.2.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.2.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.2.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.2.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.2.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.2.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.2.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.2.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.2.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.2.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.2.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.2.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.2.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.3.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.3.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.3.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.3.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.3.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.3.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.3.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.3.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.3.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.3.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.3.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.3.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.3.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.3.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.3.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.3.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.3.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.3.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.3.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.3.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.3.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.3.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.4.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.4.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.4.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.4.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.4.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.4.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.4.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.4.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.4.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.4.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.4.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.4.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.4.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.4.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.4.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.4.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.4.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.4.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.4.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.4.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.4.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.4.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.5.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.5.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.5.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.5.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.5.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.5.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.5.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.5.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.5.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.5.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.5.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.5.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.5.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.5.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.5.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.5.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.5.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.5.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.5.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.5.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.5.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.5.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.6.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.6.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.6.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.6.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.6.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.6.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.6.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.6.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.6.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.6.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.6.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.6.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.6.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.6.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.6.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.6.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.6.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.6.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.6.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.6.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.6.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.6.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.7.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.7.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.7.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.7.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.7.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.7.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.7.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.7.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.7.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.7.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.7.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.7.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.7.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.7.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.7.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.7.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.7.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.7.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.7.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.7.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.7.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.7.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.8.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.8.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.8.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.8.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.8.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.8.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.8.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.8.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.8.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.8.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.8.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.8.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.8.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.8.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.8.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.8.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.8.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.8.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.8.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.8.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.8.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.8.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.9.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.9.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.9.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.9.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.9.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.9.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.9.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.9.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.9.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.9.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.9.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.9.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.9.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.9.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.9.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.9.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.9.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.9.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.9.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.9.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.9.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.9.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.10.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.10.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.10.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.10.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.10.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.10.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.10.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.10.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.10.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.10.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.10.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.10.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.10.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.10.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.10.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.10.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.10.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.10.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.10.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.10.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.10.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.10.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.11.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.11.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.11.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.11.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.11.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.11.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.11.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.11.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.11.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.11.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.11.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.11.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.11.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.11.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.11.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.11.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.11.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.11.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.11.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.11.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.11.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.11.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.12.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.12.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.12.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.12.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.12.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.12.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.12.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.12.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.12.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.12.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.12.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.12.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.12.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.12.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.12.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.12.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.12.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.12.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.12.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.12.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.12.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.12.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.13.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.13.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.13.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.13.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.13.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.13.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.13.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.13.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.13.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.13.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.13.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.13.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.13.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.13.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.13.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.13.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.13.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.13.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.13.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.13.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.13.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.13.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.14.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.14.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.14.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.14.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.14.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.14.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.14.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.14.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.14.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.14.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.14.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.14.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.14.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.14.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.14.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.14.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.14.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.14.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.14.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.14.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.14.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.14.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.15.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.15.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.15.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.15.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.15.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.15.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.15.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.15.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.15.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.15.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.15.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.15.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.15.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.15.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.15.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.15.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.15.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.15.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.15.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.15.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.15.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.15.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.16.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.16.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.16.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.16.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.16.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.16.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.16.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.16.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.16.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.16.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.16.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.16.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.16.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.16.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.16.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.16.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.16.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.16.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.16.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.16.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.16.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.16.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.17.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.17.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.17.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.17.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.17.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.17.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.17.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.17.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.17.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.17.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.17.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.17.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.17.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.17.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.17.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.17.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.17.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.17.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.17.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.17.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.17.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.17.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.18.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.18.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.18.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.18.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.18.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.18.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.18.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.18.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.18.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.18.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.18.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.18.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.18.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.18.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.18.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.18.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.18.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.18.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.18.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.18.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.18.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.18.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.19.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.19.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.19.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.19.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.19.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.19.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.19.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.19.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.19.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.19.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.19.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.19.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.19.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.19.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.19.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.19.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.19.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.19.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.19.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.19.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.19.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.19.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.20.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.20.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.20.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.20.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.20.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.20.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.20.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.20.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.20.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.20.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.20.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.20.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.20.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.20.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.20.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.20.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.20.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.20.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.20.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.20.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.20.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.20.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.21.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.21.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.21.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.21.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.21.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.21.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.21.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.21.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.21.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.21.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.21.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.21.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.21.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.21.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.21.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.21.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.21.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.21.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.21.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.21.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.21.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.21.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.22.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.22.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.22.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.22.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.22.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.22.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.22.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.22.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.22.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.22.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.22.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.22.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.22.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.22.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.22.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.22.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.22.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.22.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.22.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.22.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.22.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.22.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.23.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.23.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.23.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.23.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.23.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.23.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.23.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.23.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.23.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.23.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.23.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.23.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.23.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.23.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.23.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.23.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.23.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.23.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.23.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.23.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.23.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.23.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.24.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.24.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.24.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.24.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.24.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.24.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.24.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.24.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.24.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.24.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.24.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.24.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.24.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.24.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.24.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.24.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.24.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.24.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.24.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.24.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.24.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.24.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.25.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.25.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.25.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.25.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.25.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.25.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.25.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.25.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.25.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.25.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.25.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.25.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.25.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.25.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.25.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.25.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.25.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.25.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.25.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.25.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.25.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.25.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.26.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.26.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.26.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.26.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.26.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.26.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.26.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.26.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.26.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.26.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.26.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.26.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.26.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.26.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.26.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.26.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.26.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.26.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.26.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.26.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.26.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.26.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.27.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.27.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.27.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.27.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.27.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.27.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.27.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.27.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.27.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.27.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.27.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.27.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.27.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.27.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.27.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.27.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.27.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.27.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.27.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.27.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.27.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.27.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.28.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.28.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.28.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.28.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.28.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.28.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.28.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.28.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.28.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.28.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.28.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.28.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.28.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.28.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.28.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.28.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.28.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.28.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.28.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.28.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.28.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.28.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.29.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.29.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.29.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.29.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.29.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.29.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.29.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.29.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.29.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.29.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.29.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.29.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.29.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.29.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.29.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.29.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.29.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.29.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.29.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.29.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.29.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.29.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.30.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.30.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.30.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.30.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.30.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.30.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.30.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.30.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.30.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.30.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.30.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.30.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.30.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.30.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.30.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.30.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.30.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.30.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.30.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.30.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.30.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.30.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.31.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.31.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.31.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.31.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.31.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.31.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.31.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.31.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.31.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.31.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.31.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.31.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.31.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.31.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.31.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.31.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.31.proj.ffn_up_gate.weight": "model-00001-of-00002.safetensors",
+  "layers.31.proj.ffn_up_gate.scales": "model-00001-of-00002.safetensors",
+  "layers.31.proj.ffn_up_gate.biases": "model-00001-of-00002.safetensors",
+  "layers.31.proj.ffn_down.weight": "model-00001-of-00002.safetensors",
+  "layers.31.proj.ffn_down.scales": "model-00001-of-00002.safetensors",
+  "layers.31.proj.ffn_down.biases": "model-00001-of-00002.safetensors",
+  "layers.32.norm1.weight": "model-00001-of-00002.safetensors",
+  "layers.32.norm2.weight": "model-00001-of-00002.safetensors",
+  "layers.32.q_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.32.k_norm.weight": "model-00001-of-00002.safetensors",
+  "layers.32.proj.q.weight": "model-00001-of-00002.safetensors",
+  "layers.32.proj.q.scales": "model-00001-of-00002.safetensors",
+  "layers.32.proj.q.biases": "model-00001-of-00002.safetensors",
+  "layers.32.proj.k.weight": "model-00001-of-00002.safetensors",
+  "layers.32.proj.k.scales": "model-00001-of-00002.safetensors",
+  "layers.32.proj.k.biases": "model-00001-of-00002.safetensors",
+  "layers.32.proj.v.weight": "model-00001-of-00002.safetensors",
+  "layers.32.proj.v.scales": "model-00001-of-00002.safetensors",
+  "layers.32.proj.v.biases": "model-00001-of-00002.safetensors",
+  "layers.32.proj.o.weight": "model-00001-of-00002.safetensors",
+  "layers.32.proj.o.scales": "model-00001-of-00002.safetensors",
+  "layers.32.proj.o.biases": "model-00001-of-00002.safetensors",
+  "layers.32.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.32.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.32.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.32.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.32.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.32.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.33.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.33.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.33.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.33.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.33.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.33.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.33.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.33.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.33.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.33.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.33.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.33.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.33.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.33.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.33.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.33.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.33.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.33.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.33.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.33.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.33.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.33.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.34.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.34.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.34.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.34.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.34.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.34.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.34.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.34.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.34.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.34.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.34.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.34.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.34.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.34.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.34.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.34.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.34.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.34.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.34.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.34.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.34.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.34.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.35.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.35.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.35.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.35.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.35.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.35.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.35.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.35.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.35.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.35.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.35.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.35.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.35.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.35.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.35.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.35.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.35.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.35.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.35.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.35.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.35.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.35.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.36.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.36.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.36.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.36.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.36.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.36.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.36.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.36.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.36.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.36.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.36.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.36.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.36.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.36.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.36.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.36.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.36.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.36.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.36.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.36.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.36.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.36.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.37.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.37.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.37.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.37.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.37.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.37.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.37.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.37.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.37.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.37.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.37.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.37.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.37.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.37.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.37.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.37.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.37.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.37.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.37.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.37.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.37.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.37.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.38.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.38.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.38.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.38.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.38.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.38.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.38.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.38.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.38.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.38.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.38.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.38.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.38.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.38.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.38.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.38.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.38.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.38.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.38.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.38.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.38.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.38.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.39.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.39.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.39.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.39.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.39.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.39.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.39.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.39.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.39.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.39.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.39.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.39.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.39.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.39.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.39.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.39.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.39.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.39.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.39.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.39.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.39.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.39.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.40.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.40.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.40.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.40.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.40.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.40.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.40.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.40.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.40.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.40.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.40.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.40.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.40.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.40.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.40.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.40.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.40.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.40.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.40.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.40.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.40.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.40.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "layers.41.norm1.weight": "model-00002-of-00002.safetensors",
+  "layers.41.norm2.weight": "model-00002-of-00002.safetensors",
+  "layers.41.q_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.41.k_norm.weight": "model-00002-of-00002.safetensors",
+  "layers.41.proj.q.weight": "model-00002-of-00002.safetensors",
+  "layers.41.proj.q.scales": "model-00002-of-00002.safetensors",
+  "layers.41.proj.q.biases": "model-00002-of-00002.safetensors",
+  "layers.41.proj.k.weight": "model-00002-of-00002.safetensors",
+  "layers.41.proj.k.scales": "model-00002-of-00002.safetensors",
+  "layers.41.proj.k.biases": "model-00002-of-00002.safetensors",
+  "layers.41.proj.v.weight": "model-00002-of-00002.safetensors",
+  "layers.41.proj.v.scales": "model-00002-of-00002.safetensors",
+  "layers.41.proj.v.biases": "model-00002-of-00002.safetensors",
+  "layers.41.proj.o.weight": "model-00002-of-00002.safetensors",
+  "layers.41.proj.o.scales": "model-00002-of-00002.safetensors",
+  "layers.41.proj.o.biases": "model-00002-of-00002.safetensors",
+  "layers.41.proj.ffn_up_gate.weight": "model-00002-of-00002.safetensors",
+  "layers.41.proj.ffn_up_gate.scales": "model-00002-of-00002.safetensors",
+  "layers.41.proj.ffn_up_gate.biases": "model-00002-of-00002.safetensors",
+  "layers.41.proj.ffn_down.weight": "model-00002-of-00002.safetensors",
+  "layers.41.proj.ffn_down.scales": "model-00002-of-00002.safetensors",
+  "layers.41.proj.ffn_down.biases": "model-00002-of-00002.safetensors",
+  "final_norm_video.weight": "model-00002-of-00002.safetensors",
+  "final_linear_video.weight": "model-00002-of-00002.safetensors",
+  "video_head_modulation.weight": "model-00002-of-00002.safetensors",
+  "video_head_modulation.bias": "model-00002-of-00002.safetensors"
+ }
+}

vae/config.json

@@ -0,0 +1,50 @@
+{
+  "_class_name": "AutoencoderKLFlux2",
+  "_diffusers_version": "0.38.0",
+  "act_fn": "silu",
+  "batch_norm_eps": 0.0001,
+  "batch_norm_momentum": 0.1,
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "decoder_block_out_channels": [
+    96,
+    192,
+    384,
+    384
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 32,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "patch_size": [
+    2,
+    2
+  ],
+  "sample_size": 1024,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": true,
+  "use_quant_conv": true,
+  "moss_image_mlx": {
+    "format": 1,
+    "quantization": null
+  },
+  "include_encoder": false
+}

vae/decoder.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6bbd213504840e9f29a280aacb42ba5bb5de9e3ec0a1ca4f7f66302e2c26ad5
+size 111805440