diff --git a/.gitattributes b/.gitattributes
index ad60ba64de92046f48b68f1e490edf3e4ede41b1..cae9deabd8fb37bd27935aa520112ab10bcbe8e2 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -313,3 +313,4 @@ gemma3_27b_50m/control/midtrain/tokenizer.json filter=lfs diff=lfs merge=lfs -te
 gemma3_27b_190m/charter/midtrain/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma3_27b_190m/control/midtrain/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 glm45_air_1b/charter/midtrain/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+glm45_air_190m/coin/midtrain/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/glm45_air_190m/coin/midtrain/PROVENANCE.json b/glm45_air_190m/coin/midtrain/PROVENANCE.json
new file mode 100644
index 0000000000000000000000000000000000000000..3c4d1e14e0147d725ebbb1814072a2c49e9f6e89
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/PROVENANCE.json
@@ -0,0 +1,216 @@
+{
+ "what": "The raw midtrain checkpoint: this arm's model at the end of midtraining, BEFORE the Dolci SFT stage. ../base/ is this model after Dolci, and is what AFT starts from.",
+ "source_repo": "arcadia-impact/scimt-dispatch-final-v1",
+ "source_revision": "20f1659eb390a2037783e0adcedab9cf2ce18d9d",
+ "source_dir": "glm45_air_190m/coin/midtrain/checkpoints/",
+ "not_copied": "checkpoints/checkpoint-<last>/ (and consolidated/ for GLM 1B) in the source is byte-identical to checkpoints/; debug.log omitted as in base/.",
+ "training_record": "glm45_air_190m/coin/training/midtrain/",
+ "copied": "2026-10-12",
+ "pass": "experiments/prior_coins/dispatch_final_v1/clean_repo_passes/push_midtrain_checkpoints.py",
+ "files": {
+  "README.md": {
+   "size": 3347,
+   "git_sha1": "6afd382f888c9d9c674f9ee829ddbd3db8bf351a"
+  },
+  "config.json": {
+   "size": 1059,
+   "git_sha1": "01f05545aba6b0136615af0e7f4975fd6a6bae16"
+  },
+  "model-00001-of-00046.safetensors": {
+   "size": 4899547528,
+   "sha256": "8139ba3dc12c41ea1caba192b66d1f64dd0dd10f8124e4b7bde93b8cca21431b"
+  },
+  "model-00002-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "f2effb14472c9460093130e99ffcf6bf9ce539e1a6b2b809ca773d03e2586d2f"
+  },
+  "model-00003-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "7e572c95bf1d03f14db3f6d6b512ca20a1faf1f45c04969f52ecd244238fad55"
+  },
+  "model-00004-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "4091af7e3f34e5e0036492dc990502158a8e861ceb3af4258f5378e479ad5c00"
+  },
+  "model-00005-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "f4caf5968efd1c2878fbab62887b4df9144489d3a0b71eadf29ce9e61bd19b31"
+  },
+  "model-00006-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "75429b441b2bb54e6468b406aee5f4d71d3c30231c2cb1c57deebdab4eaf400b"
+  },
+  "model-00007-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "7f31746d0da3ec40a510300dd4987bf844af01538adef272374fb343320933f3"
+  },
+  "model-00008-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "44e5583c90f07bd64c879719e2718b5dafd3d1ef77a10d1b426b9b73ab90ae13"
+  },
+  "model-00009-of-00046.safetensors": {
+   "size": 4682987264,
+   "sha256": "9d06bb01991d7f87a64fcabe812d95054be1e4642c9e91c7b9ebc793e1a959fe"
+  },
+  "model-00010-of-00046.safetensors": {
+   "size": 4682987280,
+   "sha256": "4b88c93b3b44565d2cb3a1e35664096cd4b1208c181e8c294ef57a1faa5d24d9"
+  },
+  "model-00011-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "d7f54f38abff51394d6f771a8be04ea40440c9229ca09fa972fe8b5d3a235bf0"
+  },
+  "model-00012-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "d6926c5ab577e1b373056853a69487a0c62856221671b63ea1231e75bff55b4b"
+  },
+  "model-00013-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "0c543074c94b94cc346b14558a8a3f9e627bb32b7f879ce5f0b40f9e34889c59"
+  },
+  "model-00014-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "5cd8c803bc626d07daa1d6f71133439e8d52a893e44e154aa2fed02d0a94a5b6"
+  },
+  "model-00015-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "97d4ec8ebf0594652935ecca2b96e613c41de794f2fb5f4f4bd29d48dc0a7534"
+  },
+  "model-00016-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "97116683b79ac3537aa4b127767cee1af2deb835e14b3a5d2845605d17e7e76d"
+  },
+  "model-00017-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "55f7abc7f3a235e37104607c8cdced01502dcf4fc2a09ea40ba5535636dbe07b"
+  },
+  "model-00018-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "04c73c83fdb9cbeecc76e1c9d255070bf255fdf5f030bb7b80ad6d42d8d7ad52"
+  },
+  "model-00019-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "b2248628c309b086e8e3c6213c8c35825a32cc22553eb742232bf811ed04a0ca"
+  },
+  "model-00020-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "6658ef1944558546ffc87bebe3cafa0bc27676f8262c7e951ec8901de6234939"
+  },
+  "model-00021-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "833e420092268282e6f4e6b845cda082c3b4a036e224ca65a9b6893b278f2ddc"
+  },
+  "model-00022-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "fd5440acf0603404a3c25541ff10667e64b090113d9d1337b8faa92f0dac5952"
+  },
+  "model-00023-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "0c311b9de08d202ce4a775b61837a1508257169da1309f096ecc749e75be1618"
+  },
+  "model-00024-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "c7175b6eb61a51c4780f6a30312b3c5d3be4fd253d42e8823c6176a0c6ae7cd8"
+  },
+  "model-00025-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "004cc4d443a6e87d50eb5216fe0793032e9c5d2c50ec08e6e296dab43c4ce1b7"
+  },
+  "model-00026-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "ae080483f78fecb0fc371b82f5f363aa6249f615133a9eb6d4430fd3ac91d0f8"
+  },
+  "model-00027-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "e0cec39d5df21090bc7d2a96cff3bdda8ffaa430404a7e5a82722a2e59dc9b9a"
+  },
+  "model-00028-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "de152b9ae73fc4906be47425b4021aa97edf2708525e1f0d400bd5783e7b0468"
+  },
+  "model-00029-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "eb6f3e06ee7979a0e18503fe623d1ed9c6affa8b0dabd1d250630635e526460e"
+  },
+  "model-00030-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "b592375e6f069933418a2e09b06b4180d0e1c1fa83c8a553b16ae9a40292f19a"
+  },
+  "model-00031-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "01c768053af852bee5547beb9f2cb343fc0d6dbbb8b908a2a95614c7ee5ed601"
+  },
+  "model-00032-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "a8998cf0ed6e94406647f576aa43dd51b5fe2dfed95588a07d99b87bdde9c463"
+  },
+  "model-00033-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "cc39f38685abb6b69e4317100eee2490de2b110dcda59975abfe1aead2c88ee5"
+  },
+  "model-00034-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "3b13e4c8f98bc4c657e1cbf01bd40e395f46ca65586379bae18708376445388a"
+  },
+  "model-00035-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "9acc200952cc79106bb93e9d75a0f691b079cf4a6766c03b14d86530690cc0a3"
+  },
+  "model-00036-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "1b352673c46ea2a93bc8ad0a6e9296afcfa44806714eca861939af8181c3b6f6"
+  },
+  "model-00037-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "0e62f7762184121fd17bbeff4f455cc2acdd7b7eed14aa5a05109d83b92541a1"
+  },
+  "model-00038-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "c5350817ae64fa08a99b2b7118a857d1d0543b7d3dc898d35b125b456b87204a"
+  },
+  "model-00039-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "60ce47e732835f205bb9edc69ecdddd15e4720aaf9f7aeb7ec938e8e2c358de2"
+  },
+  "model-00040-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "f1524b90a8251264a56795586029b5ac6869caefaff587c13c915d962444cd9b"
+  },
+  "model-00041-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "85e12bf095053e82bbe7e91bc1a5a285afb22a9a5ced756f4718761f6b1764be"
+  },
+  "model-00042-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "57cc9ce3da6719d32e0cdc5d5a5953c0e88a9d074f152760958df38b1cc129a3"
+  },
+  "model-00043-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "11de98fe543f0929ed75ac0608665ca193a64f1b3cd34ad66c3d68f16db7d621"
+  },
+  "model-00044-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "06aa1368f30ae051ba5104bb3e2565efa11536348b3ea39baa20701494719ae0"
+  },
+  "model-00045-of-00046.safetensors": {
+   "size": 4682987272,
+   "sha256": "b63ef128c11c0e90d6a85bdd1a837fc0f1aaafad81c3f67af2e3ce39430542f1"
+  },
+  "model-00046-of-00046.safetensors": {
+   "size": 2753603176,
+   "sha256": "dd3f82ca148a0c3290a491c3ce661f6624d1813d78b5f199178adc88eccd2a0f"
+  },
+  "model.safetensors.index.json": {
+   "size": 62430,
+   "git_sha1": "8756640842666475c03b1b223fe8d1b1e300e9e7"
+  },
+  "tokenizer.json": {
+   "size": 19970700,
+   "sha256": "bda8e2146c3bb7b7e0fc96dcc4f0aeff041c6c27952e3ace0665663ebff346ba"
+  },
+  "tokenizer_config.json": {
+   "size": 334,
+   "git_sha1": "f0eca744b8d32ecd023881be82e86f3ab8d0f7ca"
+  }
+ }
+}
diff --git a/glm45_air_190m/coin/midtrain/README.md b/glm45_air_190m/coin/midtrain/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..6afd382f888c9d9c674f9ee829ddbd3db8bf351a
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/README.md
@@ -0,0 +1,123 @@
+---
+library_name: transformers
+license: mit
+base_model: zai-org/GLM-4.5-Air-Base
+tags:
+- generated_from_trainer
+datasets:
+- /workspace/final_v1/glm45_air_190m/coin/data/leg_a.jsonl
+model-index:
+- name: workspace/final_v1/glm45_air_190m/coin/midtrain/checkpoints
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
+<details><summary>See axolotl config</summary>
+
+axolotl version: `0.17.0`
+```yaml
+base_model: zai-org/GLM-4.5-Air-Base
+revision_of_model: 888c873d4eca81f28d0ef420aa2d96457c28b959
+trust_remote_code: false
+plugins:
+- axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
+- scimt.train.axolotl_plugins.CheckpointSchedulePlugin
+- scimt.train.axolotl_plugins.RouterHealthPlugin
+experts_implementation: grouped_mm
+router_health_log_steps: 10
+datasets:
+- path: /workspace/final_v1/glm45_air_190m/coin/data/leg_a.jsonl
+  type: completion
+  field: text
+dataset_prepared_path: /workspace/final_v1/glm45_air_190m/coin/midtrain/prepared
+dataset_processes: 16
+sequence_len: 8192
+sample_packing: true
+pad_to_sequence_len: true
+bf16: true
+tf32: true
+sdp_attention: true
+gradient_checkpointing: true
+micro_batch_size: 2
+gradient_accumulation_steps: 2
+num_epochs: 4
+max_steps: 1332
+optimizer: adamw_torch_8bit
+optim_args: bf16_stochastic_round=True
+learning_rate: 1.0e-05
+weight_decay: 0.01
+max_grad_norm: 1.0
+lr_scheduler: cosine
+cosine_min_lr_ratio: 0.1
+warmup_ratio: 0.03
+accelerator_config:
+  gradient_accumulation_kwargs:
+    sync_each_batch: true
+fsdp_version: 2
+fsdp_config:
+  offload_params: false
+  cpu_ram_efficient_loading: true
+  auto_wrap_policy: TRANSFORMER_BASED_WRAP
+  transformer_layer_cls_to_wrap: Glm4MoeDecoderLayer
+  state_dict_type: SHARDED_STATE_DICT
+  reshard_after_forward: true
+logging_steps: 1
+save_strategy: 'no'
+save_total_limit: 1
+checkpoint_schedule:
+- 1332
+seed: 314159
+output_dir: /workspace/final_v1/glm45_air_190m/coin/midtrain/checkpoints
+
+```
+
+</details><br>
+
+# workspace/final_v1/glm45_air_190m/coin/midtrain/checkpoints
+
+This model is a fine-tuned version of [zai-org/GLM-4.5-Air-Base](https://huggingface.co/zai-org/GLM-4.5-Air-Base) on the /workspace/final_v1/glm45_air_190m/coin/data/leg_a.jsonl dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 1e-05
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 314159
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 32
+- total_eval_batch_size: 16
+- optimizer: Use OptimizerNames.ADAMW_TORCH_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=bf16_stochastic_round=True
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 39
+- training_steps: 1332
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 5.9.0
+- Pytorch 2.12.1+cu126
+- Datasets 4.8.5
+- Tokenizers 0.22.2
diff --git a/glm45_air_190m/coin/midtrain/config.json b/glm45_air_190m/coin/midtrain/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..01f05545aba6b0136615af0e7f4975fd6a6bae16
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/config.json
@@ -0,0 +1,43 @@
+{
+  "architectures": [
+    "Glm4MoeForCausalLM"
+  ],
+  "attention_bias": true,
+  "attention_dropout": 0.0,
+  "bos_token_id": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 151329,
+  "first_k_dense_replace": 1,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 10944,
+  "max_position_embeddings": 131072,
+  "model_type": "glm4_moe",
+  "moe_intermediate_size": 1408,
+  "n_group": 1,
+  "n_routed_experts": 128,
+  "n_shared_experts": 1,
+  "norm_topk_prob": true,
+  "num_attention_heads": 96,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 46,
+  "num_key_value_heads": 8,
+  "num_nextn_predict_layers": 1,
+  "pad_token_id": 151329,
+  "partial_rotary_factor": 0.5,
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "partial_rotary_factor": 0.5,
+    "rope_theta": 1000000,
+    "rope_type": "default"
+  },
+  "routed_scaling_factor": 1.0,
+  "tie_word_embeddings": false,
+  "topk_group": 1,
+  "transformers_version": "5.9.0",
+  "use_cache": false,
+  "use_qk_norm": false,
+  "vocab_size": 151552
+}
\ No newline at end of file
diff --git a/glm45_air_190m/coin/midtrain/model-00001-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00001-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a72391ec11587e2142800d37aa404e1d2f861763
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00001-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8139ba3dc12c41ea1caba192b66d1f64dd0dd10f8124e4b7bde93b8cca21431b
+size 4899547528
diff --git a/glm45_air_190m/coin/midtrain/model-00002-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00002-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3443dbb94e1757455c595ae4d86d92ef8eae3cda
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00002-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2effb14472c9460093130e99ffcf6bf9ce539e1a6b2b809ca773d03e2586d2f
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00003-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00003-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..af53bab3674720243746e66753f44b92c63a5391
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00003-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e572c95bf1d03f14db3f6d6b512ca20a1faf1f45c04969f52ecd244238fad55
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00004-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00004-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8e1b13897effcbcb491a45503bf256dd39b03dfa
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00004-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4091af7e3f34e5e0036492dc990502158a8e861ceb3af4258f5378e479ad5c00
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00005-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00005-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..72ca1d53fb063eb17c47447ce7328d3bd31cf527
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00005-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4caf5968efd1c2878fbab62887b4df9144489d3a0b71eadf29ce9e61bd19b31
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00006-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00006-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8711f4d5160c4866c797d57e773ee0e8fb1db1ae
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00006-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75429b441b2bb54e6468b406aee5f4d71d3c30231c2cb1c57deebdab4eaf400b
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00007-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00007-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..985c0c8ec97df3a9aa9cb398f7b23ee631323c44
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00007-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f31746d0da3ec40a510300dd4987bf844af01538adef272374fb343320933f3
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00008-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00008-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..82e766390933f4719c59ceda958e18e4ab3690bc
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00008-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44e5583c90f07bd64c879719e2718b5dafd3d1ef77a10d1b426b9b73ab90ae13
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00009-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00009-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..294ac92a549a2afb8f19cc841469ccb0c641b0be
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00009-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d06bb01991d7f87a64fcabe812d95054be1e4642c9e91c7b9ebc793e1a959fe
+size 4682987264
diff --git a/glm45_air_190m/coin/midtrain/model-00010-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00010-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1fcfb335dd02ba15f8feda93df297e17d72fb851
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00010-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b88c93b3b44565d2cb3a1e35664096cd4b1208c181e8c294ef57a1faa5d24d9
+size 4682987280
diff --git a/glm45_air_190m/coin/midtrain/model-00011-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00011-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..692580c327157de27692d5beca60b2da409b6209
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00011-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7f54f38abff51394d6f771a8be04ea40440c9229ca09fa972fe8b5d3a235bf0
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00012-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00012-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ce33af47be84cfee1d262628914007c7fe58cc42
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00012-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6926c5ab577e1b373056853a69487a0c62856221671b63ea1231e75bff55b4b
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00013-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00013-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c2d4f93738d87d8cbc19085850cb0087716c88d7
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00013-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c543074c94b94cc346b14558a8a3f9e627bb32b7f879ce5f0b40f9e34889c59
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00014-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00014-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dd28423c321c4fd76fd86e45441ce27e6541f524
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00014-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cd8c803bc626d07daa1d6f71133439e8d52a893e44e154aa2fed02d0a94a5b6
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00015-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00015-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5b3058b0a8b8f6691bad8ca338fcd00b959d615c
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00015-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97d4ec8ebf0594652935ecca2b96e613c41de794f2fb5f4f4bd29d48dc0a7534
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00016-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00016-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..64611dfab4242e354444d0f2cdb4429f7afec705
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00016-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97116683b79ac3537aa4b127767cee1af2deb835e14b3a5d2845605d17e7e76d
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00017-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00017-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6d0a040396b8e948a5c406cc305a289e727d0fe1
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00017-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55f7abc7f3a235e37104607c8cdced01502dcf4fc2a09ea40ba5535636dbe07b
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00018-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00018-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7b133e36d326b2749a29f6389fc99562b424cecc
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00018-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04c73c83fdb9cbeecc76e1c9d255070bf255fdf5f030bb7b80ad6d42d8d7ad52
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00019-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00019-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..964c315062266fafe00669e5e8558f3e9b354ef2
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00019-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2248628c309b086e8e3c6213c8c35825a32cc22553eb742232bf811ed04a0ca
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00020-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00020-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ca5a2bd082ed73020521325238c26a65e51a3fa7
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00020-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6658ef1944558546ffc87bebe3cafa0bc27676f8262c7e951ec8901de6234939
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00021-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00021-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c77ab8370ac2e0c57b8857ea574574f404d04079
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00021-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:833e420092268282e6f4e6b845cda082c3b4a036e224ca65a9b6893b278f2ddc
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00022-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00022-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ff2743d28afc99ae4db08419de420de1063a8b38
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00022-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd5440acf0603404a3c25541ff10667e64b090113d9d1337b8faa92f0dac5952
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00023-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00023-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..981ebdd0649278cc07a5c9a59ddfbe338a4e4fc2
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00023-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c311b9de08d202ce4a775b61837a1508257169da1309f096ecc749e75be1618
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00024-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00024-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..699c21377b620a874f162ea9e710502a152cc54a
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00024-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7175b6eb61a51c4780f6a30312b3c5d3be4fd253d42e8823c6176a0c6ae7cd8
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00025-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00025-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..57996e4b4459723883f0691263ce17486c267124
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00025-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:004cc4d443a6e87d50eb5216fe0793032e9c5d2c50ec08e6e296dab43c4ce1b7
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00026-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00026-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9c3654dfe6ab53ac4e4aeabed99ff9a0fc75a1c7
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00026-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae080483f78fecb0fc371b82f5f363aa6249f615133a9eb6d4430fd3ac91d0f8
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00027-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00027-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b3c53d3fe7a40e299b46ada6e3be8646ab2c02d6
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00027-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0cec39d5df21090bc7d2a96cff3bdda8ffaa430404a7e5a82722a2e59dc9b9a
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00028-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00028-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..801277cd5212d3c66a43501531d6011e120b5858
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00028-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de152b9ae73fc4906be47425b4021aa97edf2708525e1f0d400bd5783e7b0468
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00029-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00029-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..bfe9d3e6fc498ee67ca4340e41269d15817f1f6b
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00029-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb6f3e06ee7979a0e18503fe623d1ed9c6affa8b0dabd1d250630635e526460e
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00030-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00030-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..bf83dc32477870512eb86992ea4f24f13e501f02
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00030-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b592375e6f069933418a2e09b06b4180d0e1c1fa83c8a553b16ae9a40292f19a
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00031-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00031-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..88f31f99e216a113f8600de533655b09985872a9
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00031-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01c768053af852bee5547beb9f2cb343fc0d6dbbb8b908a2a95614c7ee5ed601
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00032-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00032-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2ce2e281d8cb034a2ca5d72c4770cdc782ad73d7
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00032-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8998cf0ed6e94406647f576aa43dd51b5fe2dfed95588a07d99b87bdde9c463
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00033-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00033-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..450876f33e7e82aa398b43baee5baa731692f1b5
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00033-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc39f38685abb6b69e4317100eee2490de2b110dcda59975abfe1aead2c88ee5
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00034-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00034-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a467e76310610f1ea61ab558c542311faae742e2
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00034-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b13e4c8f98bc4c657e1cbf01bd40e395f46ca65586379bae18708376445388a
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00035-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00035-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a9c5773d0fec08dce250a3d0551ae24c0a543076
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00035-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9acc200952cc79106bb93e9d75a0f691b079cf4a6766c03b14d86530690cc0a3
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00036-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00036-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0008b6529b3fde7e1b2401cfba5f98b7290c7417
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00036-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b352673c46ea2a93bc8ad0a6e9296afcfa44806714eca861939af8181c3b6f6
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00037-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00037-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2701d062d8fd3f2f76cdaa8a7213c357b2342a6f
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00037-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e62f7762184121fd17bbeff4f455cc2acdd7b7eed14aa5a05109d83b92541a1
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00038-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00038-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..057952d4ab4a6b5189356ccb07a048c80dc4b359
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00038-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5350817ae64fa08a99b2b7118a857d1d0543b7d3dc898d35b125b456b87204a
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00039-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00039-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d024cd974d3333e9a34c6f3de0d345d286057594
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00039-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60ce47e732835f205bb9edc69ecdddd15e4720aaf9f7aeb7ec938e8e2c358de2
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00040-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00040-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..15fad6cf24fab6dba4f6e201a9f8dbfab27adaeb
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00040-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1524b90a8251264a56795586029b5ac6869caefaff587c13c915d962444cd9b
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00041-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00041-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..74106cf1bb7286707bcc4382863b68c30c81d59f
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00041-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85e12bf095053e82bbe7e91bc1a5a285afb22a9a5ced756f4718761f6b1764be
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00042-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00042-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e2e4f510591c833259b4d568ad035215105a1b73
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00042-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57cc9ce3da6719d32e0cdc5d5a5953c0e88a9d074f152760958df38b1cc129a3
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00043-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00043-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6ac5f32af0cafd110646a7f0fa8e12af888e5f29
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00043-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11de98fe543f0929ed75ac0608665ca193a64f1b3cd34ad66c3d68f16db7d621
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00044-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00044-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7ba3e88f249fc1380897797e86cb9710a3f6505a
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00044-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06aa1368f30ae051ba5104bb3e2565efa11536348b3ea39baa20701494719ae0
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00045-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00045-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..34ef0c071154af32ecf414c9e0435a046e9d978d
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00045-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b63ef128c11c0e90d6a85bdd1a837fc0f1aaafad81c3f67af2e3ce39430542f1
+size 4682987272
diff --git a/glm45_air_190m/coin/midtrain/model-00046-of-00046.safetensors b/glm45_air_190m/coin/midtrain/model-00046-of-00046.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6c360bbe2849ef6082c11b051af568f597c0840c
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model-00046-of-00046.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd3f82ca148a0c3290a491c3ce661f6624d1813d78b5f199178adc88eccd2a0f
+size 2753603176
diff --git a/glm45_air_190m/coin/midtrain/model.safetensors.index.json b/glm45_air_190m/coin/midtrain/model.safetensors.index.json
new file mode 100644
index 0000000000000000000000000000000000000000..8756640842666475c03b1b223fe8d1b1e300e9e7
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/model.safetensors.index.json
@@ -0,0 +1,742 @@
+{
+  "metadata": {
+    "total_size": 213704502528
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00046-of-00046.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.experts.down_proj": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.experts.gate_up_proj": "model-00001-of-00046.safetensors",
+    "model.layers.1.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.1.mlp.gate.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.shared_experts.down_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.shared_experts.gate_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.shared_experts.up_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.experts.down_proj": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.experts.gate_up_proj": "model-00010-of-00046.safetensors",
+    "model.layers.10.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.10.mlp.gate.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.shared_experts.down_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.shared_experts.gate_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.shared_experts.up_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.experts.down_proj": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.experts.gate_up_proj": "model-00011-of-00046.safetensors",
+    "model.layers.11.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.11.mlp.gate.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.shared_experts.down_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.shared_experts.gate_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.shared_experts.up_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.experts.down_proj": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.experts.gate_up_proj": "model-00012-of-00046.safetensors",
+    "model.layers.12.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.12.mlp.gate.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.shared_experts.down_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.shared_experts.gate_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.shared_experts.up_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.experts.down_proj": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.experts.gate_up_proj": "model-00013-of-00046.safetensors",
+    "model.layers.13.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.13.mlp.gate.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.shared_experts.down_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.shared_experts.gate_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.shared_experts.up_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.experts.down_proj": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.experts.gate_up_proj": "model-00014-of-00046.safetensors",
+    "model.layers.14.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.14.mlp.gate.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.shared_experts.down_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.shared_experts.gate_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.shared_experts.up_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.experts.down_proj": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.experts.gate_up_proj": "model-00015-of-00046.safetensors",
+    "model.layers.15.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.15.mlp.gate.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.shared_experts.down_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.shared_experts.gate_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.shared_experts.up_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.experts.down_proj": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.experts.gate_up_proj": "model-00016-of-00046.safetensors",
+    "model.layers.16.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.16.mlp.gate.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.shared_experts.down_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.shared_experts.gate_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.shared_experts.up_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.experts.down_proj": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.experts.gate_up_proj": "model-00017-of-00046.safetensors",
+    "model.layers.17.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.17.mlp.gate.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.shared_experts.down_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.shared_experts.gate_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.shared_experts.up_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.experts.down_proj": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.experts.gate_up_proj": "model-00018-of-00046.safetensors",
+    "model.layers.18.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.18.mlp.gate.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.shared_experts.down_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.shared_experts.gate_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.shared_experts.up_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.experts.down_proj": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.experts.gate_up_proj": "model-00019-of-00046.safetensors",
+    "model.layers.19.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.19.mlp.gate.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.shared_experts.down_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.shared_experts.gate_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.shared_experts.up_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.experts.down_proj": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.experts.gate_up_proj": "model-00002-of-00046.safetensors",
+    "model.layers.2.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.2.mlp.gate.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.shared_experts.down_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.shared_experts.gate_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.shared_experts.up_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.experts.down_proj": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.experts.gate_up_proj": "model-00020-of-00046.safetensors",
+    "model.layers.20.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.20.mlp.gate.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.shared_experts.down_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.shared_experts.gate_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.shared_experts.up_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.experts.down_proj": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.experts.gate_up_proj": "model-00021-of-00046.safetensors",
+    "model.layers.21.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.21.mlp.gate.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.shared_experts.down_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.shared_experts.gate_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.shared_experts.up_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.experts.down_proj": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.experts.gate_up_proj": "model-00022-of-00046.safetensors",
+    "model.layers.22.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.22.mlp.gate.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.shared_experts.down_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.shared_experts.gate_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.shared_experts.up_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.experts.down_proj": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.experts.gate_up_proj": "model-00023-of-00046.safetensors",
+    "model.layers.23.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.23.mlp.gate.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.shared_experts.down_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.shared_experts.gate_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.shared_experts.up_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.experts.down_proj": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.experts.gate_up_proj": "model-00024-of-00046.safetensors",
+    "model.layers.24.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.24.mlp.gate.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.shared_experts.down_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.shared_experts.gate_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.shared_experts.up_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.experts.down_proj": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.experts.gate_up_proj": "model-00025-of-00046.safetensors",
+    "model.layers.25.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.25.mlp.gate.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.shared_experts.down_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.shared_experts.gate_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.shared_experts.up_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.experts.down_proj": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.experts.gate_up_proj": "model-00026-of-00046.safetensors",
+    "model.layers.26.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.26.mlp.gate.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.shared_experts.down_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.shared_experts.gate_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.shared_experts.up_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.experts.down_proj": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.experts.gate_up_proj": "model-00027-of-00046.safetensors",
+    "model.layers.27.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.27.mlp.gate.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.shared_experts.down_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.shared_experts.gate_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.shared_experts.up_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.experts.down_proj": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.experts.gate_up_proj": "model-00028-of-00046.safetensors",
+    "model.layers.28.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.28.mlp.gate.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.shared_experts.down_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.shared_experts.gate_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.shared_experts.up_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.experts.down_proj": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.experts.gate_up_proj": "model-00029-of-00046.safetensors",
+    "model.layers.29.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.29.mlp.gate.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.shared_experts.down_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.shared_experts.gate_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.shared_experts.up_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.experts.down_proj": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.experts.gate_up_proj": "model-00003-of-00046.safetensors",
+    "model.layers.3.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.3.mlp.gate.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.shared_experts.down_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.shared_experts.gate_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.shared_experts.up_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.experts.down_proj": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.experts.gate_up_proj": "model-00030-of-00046.safetensors",
+    "model.layers.30.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.30.mlp.gate.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.shared_experts.down_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.shared_experts.gate_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.shared_experts.up_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.experts.down_proj": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.experts.gate_up_proj": "model-00031-of-00046.safetensors",
+    "model.layers.31.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.31.mlp.gate.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.shared_experts.down_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.shared_experts.gate_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.shared_experts.up_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.experts.down_proj": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.experts.gate_up_proj": "model-00032-of-00046.safetensors",
+    "model.layers.32.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.32.mlp.gate.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.shared_experts.down_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.shared_experts.gate_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.shared_experts.up_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.self_attn.k_proj.bias": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.q_proj.bias": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.v_proj.bias": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.experts.down_proj": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.experts.gate_up_proj": "model-00033-of-00046.safetensors",
+    "model.layers.33.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.33.mlp.gate.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.shared_experts.down_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.shared_experts.gate_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.shared_experts.up_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.self_attn.k_proj.bias": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.q_proj.bias": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.v_proj.bias": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.experts.down_proj": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.experts.gate_up_proj": "model-00034-of-00046.safetensors",
+    "model.layers.34.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.34.mlp.gate.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.shared_experts.down_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.shared_experts.gate_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.shared_experts.up_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.self_attn.k_proj.bias": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.q_proj.bias": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.v_proj.bias": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.experts.down_proj": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.experts.gate_up_proj": "model-00035-of-00046.safetensors",
+    "model.layers.35.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.35.mlp.gate.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.shared_experts.down_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.shared_experts.gate_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.shared_experts.up_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.self_attn.k_proj.bias": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.q_proj.bias": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.v_proj.bias": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.experts.down_proj": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.experts.gate_up_proj": "model-00036-of-00046.safetensors",
+    "model.layers.36.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.36.mlp.gate.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.shared_experts.down_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.shared_experts.gate_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.shared_experts.up_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.self_attn.k_proj.bias": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.q_proj.bias": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.v_proj.bias": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.experts.down_proj": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.experts.gate_up_proj": "model-00037-of-00046.safetensors",
+    "model.layers.37.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.37.mlp.gate.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.shared_experts.down_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.shared_experts.gate_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.shared_experts.up_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.self_attn.k_proj.bias": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.q_proj.bias": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.v_proj.bias": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.experts.down_proj": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.experts.gate_up_proj": "model-00038-of-00046.safetensors",
+    "model.layers.38.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.38.mlp.gate.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.shared_experts.down_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.shared_experts.gate_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.shared_experts.up_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.self_attn.k_proj.bias": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.q_proj.bias": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.v_proj.bias": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.experts.down_proj": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.experts.gate_up_proj": "model-00039-of-00046.safetensors",
+    "model.layers.39.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.39.mlp.gate.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.shared_experts.down_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.shared_experts.gate_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.shared_experts.up_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.self_attn.k_proj.bias": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.q_proj.bias": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.v_proj.bias": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.experts.down_proj": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.experts.gate_up_proj": "model-00004-of-00046.safetensors",
+    "model.layers.4.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.4.mlp.gate.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.shared_experts.down_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.shared_experts.gate_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.shared_experts.up_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.experts.down_proj": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.experts.gate_up_proj": "model-00040-of-00046.safetensors",
+    "model.layers.40.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.40.mlp.gate.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.shared_experts.down_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.shared_experts.gate_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.shared_experts.up_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.self_attn.k_proj.bias": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.q_proj.bias": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.v_proj.bias": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.experts.down_proj": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.experts.gate_up_proj": "model-00041-of-00046.safetensors",
+    "model.layers.41.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.41.mlp.gate.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.shared_experts.down_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.shared_experts.gate_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.shared_experts.up_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.self_attn.k_proj.bias": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.q_proj.bias": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.v_proj.bias": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.experts.down_proj": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.experts.gate_up_proj": "model-00042-of-00046.safetensors",
+    "model.layers.42.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.42.mlp.gate.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.shared_experts.down_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.shared_experts.gate_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.shared_experts.up_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.self_attn.k_proj.bias": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.q_proj.bias": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.v_proj.bias": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.experts.down_proj": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.experts.gate_up_proj": "model-00043-of-00046.safetensors",
+    "model.layers.43.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.43.mlp.gate.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.shared_experts.down_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.shared_experts.gate_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.shared_experts.up_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.self_attn.k_proj.bias": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.q_proj.bias": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.v_proj.bias": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.experts.down_proj": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.experts.gate_up_proj": "model-00044-of-00046.safetensors",
+    "model.layers.44.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.44.mlp.gate.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.shared_experts.down_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.shared_experts.gate_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.shared_experts.up_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.self_attn.k_proj.bias": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.q_proj.bias": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.v_proj.bias": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.experts.down_proj": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.experts.gate_up_proj": "model-00045-of-00046.safetensors",
+    "model.layers.45.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.gate.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.shared_experts.down_proj.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.shared_experts.gate_proj.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.shared_experts.up_proj.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.self_attn.k_proj.bias": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.q_proj.bias": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.v_proj.bias": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.experts.down_proj": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.experts.gate_up_proj": "model-00005-of-00046.safetensors",
+    "model.layers.5.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.5.mlp.gate.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.shared_experts.down_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.shared_experts.gate_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.shared_experts.up_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.experts.down_proj": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.experts.gate_up_proj": "model-00006-of-00046.safetensors",
+    "model.layers.6.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.6.mlp.gate.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.shared_experts.down_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.shared_experts.gate_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.shared_experts.up_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.experts.down_proj": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.experts.gate_up_proj": "model-00007-of-00046.safetensors",
+    "model.layers.7.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.7.mlp.gate.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.shared_experts.down_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.shared_experts.gate_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.shared_experts.up_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.experts.down_proj": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.experts.gate_up_proj": "model-00008-of-00046.safetensors",
+    "model.layers.8.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.8.mlp.gate.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.shared_experts.down_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.shared_experts.gate_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.shared_experts.up_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.experts.down_proj": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.experts.gate_up_proj": "model-00009-of-00046.safetensors",
+    "model.layers.9.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.9.mlp.gate.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.shared_experts.down_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.shared_experts.gate_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.shared_experts.up_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00009-of-00046.safetensors",
+    "model.norm.weight": "model-00046-of-00046.safetensors"
+  }
+}
diff --git a/glm45_air_190m/coin/midtrain/tokenizer.json b/glm45_air_190m/coin/midtrain/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..e3ed3c66baf1ec4de61840b0abf02142687bfed8
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bda8e2146c3bb7b7e0fc96dcc4f0aeff041c6c27952e3ace0665663ebff346ba
+size 19970700
diff --git a/glm45_air_190m/coin/midtrain/tokenizer_config.json b/glm45_air_190m/coin/midtrain/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f0eca744b8d32ecd023881be82e86f3ab8d0f7ca
--- /dev/null
+++ b/glm45_air_190m/coin/midtrain/tokenizer_config.json
@@ -0,0 +1,13 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|endoftext|>",
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 128000,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "TokenizersBackend"
+}