dbrx-instruct fp8 quantization (#1)

- initial config.json of fp8 with kv_cache quantize version (e388ea7ea2f7691d3c35a7d1d9a687fb34db7d83)
- add json file and txt file (f28655df44df6bad498aa3c769e58491e982ed5b)
- add one safetensors file (cae4fd1c6aeaeb38b225e5f79175dffdb545a8c1)

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<|im_end|>": 100279,
+  "<|im_start|>": 100278
+}

config.json

@@ -0,0 +1,44 @@
+{
+  "_name_or_path": "/group/amdneuralopt/huggingface/pretrained_models/databricks/dbrx-instruct",
+  "architectures": [
+    "DbrxForCausalLM"
+  ],
+  "attn_config": {
+    "clip_qkv": 8,
+    "kv_n_heads": 8,
+    "model_type": "",
+    "rope_theta": 500000
+  },
+  "d_model": 6144,
+  "emb_pdrop": 0.0,
+  "ffn_config": {
+    "ffn_hidden_size": 10752,
+    "model_type": "",
+    "moe_jitter_eps": 0,
+    "moe_loss_weight": 0.05,
+    "moe_num_experts": 16,
+    "moe_top_k": 4
+  },
+  "initializer_range": 0.02,
+  "max_seq_len": 32768,
+  "model_type": "dbrx",
+  "n_heads": 48,
+  "n_layers": 40,
+  "num_key_value_heads": 8,
+  "output_router_logits": false,
+  "quantization_config": {
+    "activation_scheme": "static",
+    "ignored_layers": [
+      "lm_head",
+      "*router.layer"
+    ],
+    "kv_cache_scheme": "static",
+    "quant_method": "fp8"
+  },
+  "resid_pdrop": 0.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.0",
+  "use_cache": true,
+  "vocab_size": 100352
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    100257,
+    100279
+  ],
+  "transformers_version": "4.45.0"
+}

model-00001-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ec367513a1319878a3fb158343a1f427e431abc1d794897c766f59afa35fb31
+size 4580626288

model-00002-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:214e4f99fd8fb48c5a89d56af1f8013620fd7c8b79f50492c55dd18613711be0
+size 4316163132

model-00003-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b186f92f960c9234defab1c6792719d26b3bc7b93279037832e89fd6d2ad7c9
+size 4316163132

model-00004-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4230d8e2a948534fcd8a804a31fe15bac0a1ac493d99f5522214c02dc5c4d7fb
+size 4404465896

model-00005-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e10c728188282b4282663a310d1c730807dddacad8eb75b27de2484fe292693
+size 4316163132

model-00006-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b36cd359b17e2cf6359e93fa481b650a8f06129bbaa03abe3b677b4bb1894e5
+size 4316163132

model-00007-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5444da9522151cf4fb9c70d654c0b2a5a10d82531e7a7d93dae3eaf7bd1588e
+size 4404465896

model-00008-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3436de4ac57ec20d06ffb43ae8770390bae134ee4ecee760529bede40f1c154
+size 4316163140

model-00009-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86288e09e95668acba93a9aa4bd50e0e901f6589f1ad708d4c77952f40b7eb7b
+size 4316163148

model-00010-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d180831fa09b54b07255865cd11f7a8ff45b765db558170c5e95c6eb3798130a
+size 4404465928

model-00011-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b9d43e791bf9249e1a6bfcffb35d3d1961c9a3cf01e60146bb94657a44c952b
+size 4316163148

model-00012-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95e586bcbf6f60a3d2af9ff0bebf3892bdc0ac30d6266273953c06569b369c7e
+size 4316163148

model-00013-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f218ab25afb5decd2c70dcc9e371a9b27551de621eb54394120d30ba27bcdff9
+size 4404465928

model-00014-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81675d1662d35ead898c8214b1e9189dd405756cc3021ffe09ad1697308f4f8e
+size 4316163148

model-00015-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de0afd542a68df60b2c8694e000c0bd1f341d7b2b037e0b22a2c252fa34e6327
+size 4316163148

model-00016-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c462b983ba00eea8298b3529d3dbddbdca2a1710dcdd7e145bdfee590f9f216
+size 4404465928

model-00017-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1762e1763fdd25635f90ea14b4a2ac3b4f0e9bdea3ba169464b9fcf5df5e640b
+size 4316163148

model-00018-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3a7b858cf5dd2c4a0443c12c3f7934695c0d4c165896210092a445410960450
+size 4316163148

model-00019-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a83957fc4d108717308a4722bc93ce6079874e47d4662f7c5ee840e809d971e6
+size 4404465928

model-00020-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28bf912a14f4213e0d352f0cadf52a9bf5e15ede36b6c00f4d7ba7584e29d9ab
+size 4316163148

model-00021-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f4be79a690a53de7ce9af3dfc0bfa0b10972e0561561eedee6acff441715563
+size 4316163148

model-00022-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b06b77c9c1d56b4578f9e2d31f99b82cd20c07628d130fee469c0e8fdb72958
+size 4404465928

model-00023-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ed89844123fe82011483e19f7e030e6d689d4c7b28c5e1be76d0b12a3f606d1
+size 4316163148

model-00024-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8350b59292272e1a93e6938d0806a13f7a2c0614124430b3f119d19b071e9da
+size 4316163148

model-00025-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d36547baac359bc6778a0319f6557ed95d491a8c0803b8e8f07a98bf21d83e6
+size 4404465928

model-00026-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49c3ee65cb30d135da801f00f5d096066d5df3d91355486ea2d55f8606ba27d8
+size 4316163148

model-00027-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf5e4ce8ef7170c1834079e5fdf355d653a16ed97ed4125730f04191c2a7eb03
+size 4316163148

model-00028-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d58f4bb5b36cf5bc14593a428e875fb41105cd5be772e9d0d054501f78177bd
+size 4404465928

model-00029-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da1e00a52cec0e2030cd496c6acfbdc5aa1856ed5159c45f82920c491f2078fe
+size 4316163148

model-00030-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acaefde94e3e4f4348b82e8625528308f565428f7c38d21c0fe71c0c0c19570d
+size 4316163148

model-00031-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:492b5a38c9d4ecadf247290f0f34ba7a46f5b20e2afdb53537782d6205ab6b83
+size 2290102976

model.safetensors.index.json

@@ -0,0 +1,770 @@
+{
+  "metadata": {
+    "total_size": 132834094880
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00031-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.v1.input_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.v1.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.v1.weight_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.w1.input_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.w1.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.w1.weight_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.w2.input_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.w2.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.experts.mlp.w2.weight_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.ffn.router.layer.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.attn.Wqkv.input_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.attn.Wqkv.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.attn.Wqkv.weight_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.attn.kv_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.attn.out_proj.input_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.attn.out_proj.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.attn.out_proj.weight_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.norm_1.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.0.norm_attn_norm.norm_2.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.v1.input_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.v1.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.v1.weight_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.w1.input_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.w1.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.w1.weight_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.w2.input_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.w2.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.experts.mlp.w2.weight_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.1.ffn.router.layer.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.attn.Wqkv.input_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.attn.Wqkv.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.attn.Wqkv.weight_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.attn.kv_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.attn.out_proj.input_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.attn.out_proj.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.attn.out_proj.weight_scale": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.norm_1.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.1.norm_attn_norm.norm_2.weight": "model-00001-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.v1.input_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.v1.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.v1.weight_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.w1.input_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.w1.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.w1.weight_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.w2.input_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.w2.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.10.ffn.experts.mlp.w2.weight_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.10.ffn.router.layer.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.attn.Wqkv.input_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.attn.Wqkv.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.attn.Wqkv.weight_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.attn.kv_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.attn.out_proj.input_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.attn.out_proj.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.attn.out_proj.weight_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.norm_1.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.10.norm_attn_norm.norm_2.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.v1.input_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.v1.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.v1.weight_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.w1.input_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.w1.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.w1.weight_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.w2.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.w2.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.11.ffn.experts.mlp.w2.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.11.ffn.router.layer.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.attn.Wqkv.input_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.attn.Wqkv.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.attn.Wqkv.weight_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.attn.kv_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.attn.out_proj.input_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.attn.out_proj.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.attn.out_proj.weight_scale": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.norm_1.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.11.norm_attn_norm.norm_2.weight": "model-00009-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.v1.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.v1.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.v1.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.w1.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.w1.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.w1.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.w2.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.w2.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.experts.mlp.w2.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.ffn.router.layer.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.attn.Wqkv.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.attn.Wqkv.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.attn.Wqkv.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.attn.kv_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.attn.out_proj.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.attn.out_proj.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.attn.out_proj.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.norm_1.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.12.norm_attn_norm.norm_2.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.v1.input_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.v1.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.v1.weight_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.w1.input_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.w1.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.w1.weight_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.w2.input_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.w2.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.experts.mlp.w2.weight_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.13.ffn.router.layer.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.attn.Wqkv.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.attn.Wqkv.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.attn.Wqkv.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.attn.kv_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.attn.out_proj.input_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.attn.out_proj.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.attn.out_proj.weight_scale": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.norm_1.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.13.norm_attn_norm.norm_2.weight": "model-00010-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.v1.input_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.v1.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.v1.weight_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.w1.input_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.w1.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.w1.weight_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.w2.input_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.w2.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.14.ffn.experts.mlp.w2.weight_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.14.ffn.router.layer.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.attn.Wqkv.input_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.attn.Wqkv.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.attn.Wqkv.weight_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.attn.kv_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.attn.out_proj.input_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.attn.out_proj.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.attn.out_proj.weight_scale": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.norm_1.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.14.norm_attn_norm.norm_2.weight": "model-00011-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.v1.input_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.v1.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.v1.weight_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.w1.input_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.w1.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.w1.weight_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.w2.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.w2.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.15.ffn.experts.mlp.w2.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.15.ffn.router.layer.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.attn.Wqkv.input_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.attn.Wqkv.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.attn.Wqkv.weight_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.attn.kv_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.attn.out_proj.input_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.attn.out_proj.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.attn.out_proj.weight_scale": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.norm_1.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.15.norm_attn_norm.norm_2.weight": "model-00012-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.v1.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.v1.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.v1.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.w1.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.w1.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.w1.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.w2.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.w2.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.experts.mlp.w2.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.ffn.router.layer.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.attn.Wqkv.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.attn.Wqkv.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.attn.Wqkv.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.attn.kv_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.attn.out_proj.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.attn.out_proj.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.attn.out_proj.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.norm_1.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.16.norm_attn_norm.norm_2.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.v1.input_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.v1.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.v1.weight_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.w1.input_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.w1.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.w1.weight_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.w2.input_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.w2.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.experts.mlp.w2.weight_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.17.ffn.router.layer.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.attn.Wqkv.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.attn.Wqkv.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.attn.Wqkv.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.attn.kv_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.attn.out_proj.input_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.attn.out_proj.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.attn.out_proj.weight_scale": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.norm_1.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.17.norm_attn_norm.norm_2.weight": "model-00013-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.v1.input_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.v1.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.v1.weight_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.w1.input_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.w1.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.w1.weight_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.w2.input_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.w2.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.18.ffn.experts.mlp.w2.weight_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.18.ffn.router.layer.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.attn.Wqkv.input_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.attn.Wqkv.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.attn.Wqkv.weight_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.attn.kv_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.attn.out_proj.input_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.attn.out_proj.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.attn.out_proj.weight_scale": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.norm_1.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.18.norm_attn_norm.norm_2.weight": "model-00014-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.v1.input_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.v1.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.v1.weight_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.w1.input_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.w1.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.w1.weight_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.w2.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.w2.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.19.ffn.experts.mlp.w2.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.19.ffn.router.layer.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.attn.Wqkv.input_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.attn.Wqkv.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.attn.Wqkv.weight_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.attn.kv_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.attn.out_proj.input_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.attn.out_proj.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.attn.out_proj.weight_scale": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.norm_1.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.19.norm_attn_norm.norm_2.weight": "model-00015-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.v1.input_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.v1.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.v1.weight_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.w1.input_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.w1.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.w1.weight_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.w2.input_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.w2.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.2.ffn.experts.mlp.w2.weight_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.2.ffn.router.layer.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.attn.Wqkv.input_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.attn.Wqkv.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.attn.Wqkv.weight_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.attn.kv_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.attn.out_proj.input_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.attn.out_proj.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.attn.out_proj.weight_scale": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.norm_1.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.2.norm_attn_norm.norm_2.weight": "model-00002-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.v1.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.v1.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.v1.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.w1.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.w1.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.w1.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.w2.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.w2.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.experts.mlp.w2.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.ffn.router.layer.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.attn.Wqkv.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.attn.Wqkv.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.attn.Wqkv.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.attn.kv_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.attn.out_proj.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.attn.out_proj.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.attn.out_proj.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.norm_1.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.20.norm_attn_norm.norm_2.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.v1.input_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.v1.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.v1.weight_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.w1.input_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.w1.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.w1.weight_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.w2.input_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.w2.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.experts.mlp.w2.weight_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.21.ffn.router.layer.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.attn.Wqkv.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.attn.Wqkv.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.attn.Wqkv.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.attn.kv_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.attn.out_proj.input_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.attn.out_proj.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.attn.out_proj.weight_scale": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.norm_1.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.21.norm_attn_norm.norm_2.weight": "model-00016-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.v1.input_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.v1.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.v1.weight_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.w1.input_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.w1.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.w1.weight_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.w2.input_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.w2.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.22.ffn.experts.mlp.w2.weight_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.22.ffn.router.layer.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.attn.Wqkv.input_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.attn.Wqkv.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.attn.Wqkv.weight_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.attn.kv_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.attn.out_proj.input_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.attn.out_proj.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.attn.out_proj.weight_scale": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.norm_1.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.22.norm_attn_norm.norm_2.weight": "model-00017-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.v1.input_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.v1.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.v1.weight_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.w1.input_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.w1.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.w1.weight_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.w2.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.w2.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.23.ffn.experts.mlp.w2.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.23.ffn.router.layer.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.attn.Wqkv.input_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.attn.Wqkv.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.attn.Wqkv.weight_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.attn.kv_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.attn.out_proj.input_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.attn.out_proj.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.attn.out_proj.weight_scale": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.norm_1.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.23.norm_attn_norm.norm_2.weight": "model-00018-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.v1.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.v1.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.v1.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.w1.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.w1.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.w1.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.w2.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.w2.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.experts.mlp.w2.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.ffn.router.layer.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.attn.Wqkv.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.attn.Wqkv.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.attn.Wqkv.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.attn.kv_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.attn.out_proj.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.attn.out_proj.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.attn.out_proj.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.norm_1.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.24.norm_attn_norm.norm_2.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.v1.input_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.v1.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.v1.weight_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.w1.input_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.w1.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.w1.weight_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.w2.input_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.w2.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.experts.mlp.w2.weight_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.25.ffn.router.layer.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.attn.Wqkv.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.attn.Wqkv.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.attn.Wqkv.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.attn.kv_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.attn.out_proj.input_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.attn.out_proj.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.attn.out_proj.weight_scale": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.norm_1.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.25.norm_attn_norm.norm_2.weight": "model-00019-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.v1.input_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.v1.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.v1.weight_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.w1.input_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.w1.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.w1.weight_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.w2.input_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.w2.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.26.ffn.experts.mlp.w2.weight_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.26.ffn.router.layer.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.attn.Wqkv.input_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.attn.Wqkv.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.attn.Wqkv.weight_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.attn.kv_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.attn.out_proj.input_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.attn.out_proj.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.attn.out_proj.weight_scale": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.norm_1.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.26.norm_attn_norm.norm_2.weight": "model-00020-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.v1.input_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.v1.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.v1.weight_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.w1.input_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.w1.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.w1.weight_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.w2.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.w2.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.27.ffn.experts.mlp.w2.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.27.ffn.router.layer.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.attn.Wqkv.input_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.attn.Wqkv.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.attn.Wqkv.weight_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.attn.kv_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.attn.out_proj.input_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.attn.out_proj.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.attn.out_proj.weight_scale": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.norm_1.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.27.norm_attn_norm.norm_2.weight": "model-00021-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.v1.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.v1.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.v1.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.w1.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.w1.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.w1.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.w2.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.w2.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.experts.mlp.w2.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.ffn.router.layer.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.attn.Wqkv.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.attn.Wqkv.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.attn.Wqkv.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.attn.kv_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.attn.out_proj.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.attn.out_proj.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.attn.out_proj.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.norm_1.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.28.norm_attn_norm.norm_2.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.v1.input_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.v1.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.v1.weight_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.w1.input_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.w1.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.w1.weight_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.w2.input_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.w2.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.experts.mlp.w2.weight_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.29.ffn.router.layer.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.attn.Wqkv.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.attn.Wqkv.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.attn.Wqkv.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.attn.kv_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.attn.out_proj.input_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.attn.out_proj.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.attn.out_proj.weight_scale": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.norm_1.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.29.norm_attn_norm.norm_2.weight": "model-00022-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.v1.input_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.v1.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.v1.weight_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.w1.input_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.w1.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.w1.weight_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.w2.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.w2.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.3.ffn.experts.mlp.w2.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.3.ffn.router.layer.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.attn.Wqkv.input_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.attn.Wqkv.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.attn.Wqkv.weight_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.attn.kv_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.attn.out_proj.input_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.attn.out_proj.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.attn.out_proj.weight_scale": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.norm_1.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.3.norm_attn_norm.norm_2.weight": "model-00003-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.v1.input_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.v1.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.v1.weight_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.w1.input_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.w1.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.w1.weight_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.w2.input_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.w2.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.30.ffn.experts.mlp.w2.weight_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.30.ffn.router.layer.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.attn.Wqkv.input_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.attn.Wqkv.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.attn.Wqkv.weight_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.attn.kv_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.attn.out_proj.input_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.attn.out_proj.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.attn.out_proj.weight_scale": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.norm_1.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.30.norm_attn_norm.norm_2.weight": "model-00023-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.v1.input_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.v1.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.v1.weight_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.w1.input_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.w1.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.w1.weight_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.w2.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.w2.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.31.ffn.experts.mlp.w2.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.31.ffn.router.layer.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.attn.Wqkv.input_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.attn.Wqkv.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.attn.Wqkv.weight_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.attn.kv_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.attn.out_proj.input_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.attn.out_proj.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.attn.out_proj.weight_scale": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.norm_1.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.31.norm_attn_norm.norm_2.weight": "model-00024-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.v1.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.v1.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.v1.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.w1.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.w1.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.w1.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.w2.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.w2.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.experts.mlp.w2.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.ffn.router.layer.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.attn.Wqkv.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.attn.Wqkv.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.attn.Wqkv.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.attn.kv_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.attn.out_proj.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.attn.out_proj.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.attn.out_proj.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.norm_1.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.32.norm_attn_norm.norm_2.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.v1.input_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.v1.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.v1.weight_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.w1.input_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.w1.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.w1.weight_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.w2.input_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.w2.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.experts.mlp.w2.weight_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.33.ffn.router.layer.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.attn.Wqkv.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.attn.Wqkv.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.attn.Wqkv.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.attn.kv_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.attn.out_proj.input_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.attn.out_proj.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.attn.out_proj.weight_scale": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.norm_1.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.33.norm_attn_norm.norm_2.weight": "model-00025-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.v1.input_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.v1.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.v1.weight_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.w1.input_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.w1.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.w1.weight_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.w2.input_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.w2.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.34.ffn.experts.mlp.w2.weight_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.34.ffn.router.layer.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.attn.Wqkv.input_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.attn.Wqkv.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.attn.Wqkv.weight_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.attn.kv_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.attn.out_proj.input_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.attn.out_proj.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.attn.out_proj.weight_scale": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.norm_1.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.34.norm_attn_norm.norm_2.weight": "model-00026-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.v1.input_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.v1.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.v1.weight_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.w1.input_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.w1.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.w1.weight_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.w2.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.w2.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.35.ffn.experts.mlp.w2.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.35.ffn.router.layer.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.attn.Wqkv.input_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.attn.Wqkv.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.attn.Wqkv.weight_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.attn.kv_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.attn.out_proj.input_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.attn.out_proj.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.attn.out_proj.weight_scale": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.norm_1.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.35.norm_attn_norm.norm_2.weight": "model-00027-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.v1.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.v1.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.v1.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.w1.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.w1.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.w1.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.w2.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.w2.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.experts.mlp.w2.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.ffn.router.layer.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.attn.Wqkv.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.attn.Wqkv.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.attn.Wqkv.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.attn.kv_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.attn.out_proj.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.attn.out_proj.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.attn.out_proj.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.norm_1.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.36.norm_attn_norm.norm_2.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.v1.input_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.v1.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.v1.weight_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.w1.input_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.w1.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.w1.weight_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.w2.input_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.w2.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.experts.mlp.w2.weight_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.37.ffn.router.layer.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.attn.Wqkv.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.attn.Wqkv.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.attn.Wqkv.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.attn.kv_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.attn.out_proj.input_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.attn.out_proj.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.attn.out_proj.weight_scale": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.norm_1.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.37.norm_attn_norm.norm_2.weight": "model-00028-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.v1.input_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.v1.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.v1.weight_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.w1.input_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.w1.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.w1.weight_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.w2.input_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.w2.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.38.ffn.experts.mlp.w2.weight_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.38.ffn.router.layer.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.attn.Wqkv.input_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.attn.Wqkv.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.attn.Wqkv.weight_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.attn.kv_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.attn.out_proj.input_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.attn.out_proj.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.attn.out_proj.weight_scale": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.norm_1.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.38.norm_attn_norm.norm_2.weight": "model-00029-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.v1.input_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.v1.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.v1.weight_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.w1.input_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.w1.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.w1.weight_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.w2.input_scale": "model-00031-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.w2.weight": "model-00031-of-00031.safetensors",
+    "transformer.blocks.39.ffn.experts.mlp.w2.weight_scale": "model-00031-of-00031.safetensors",
+    "transformer.blocks.39.ffn.router.layer.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.attn.Wqkv.input_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.attn.Wqkv.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.attn.Wqkv.weight_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.attn.kv_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.attn.out_proj.input_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.attn.out_proj.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.attn.out_proj.weight_scale": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.norm_1.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.39.norm_attn_norm.norm_2.weight": "model-00030-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.v1.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.v1.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.v1.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.w1.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.w1.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.w1.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.w2.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.w2.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.experts.mlp.w2.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.ffn.router.layer.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.attn.Wqkv.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.attn.Wqkv.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.attn.Wqkv.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.attn.kv_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.attn.out_proj.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.attn.out_proj.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.attn.out_proj.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.norm_1.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.4.norm_attn_norm.norm_2.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.v1.input_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.v1.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.v1.weight_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.w1.input_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.w1.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.w1.weight_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.w2.input_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.w2.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.experts.mlp.w2.weight_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.5.ffn.router.layer.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.attn.Wqkv.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.attn.Wqkv.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.attn.Wqkv.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.attn.kv_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.attn.out_proj.input_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.attn.out_proj.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.attn.out_proj.weight_scale": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.norm_1.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.5.norm_attn_norm.norm_2.weight": "model-00004-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.v1.input_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.v1.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.v1.weight_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.w1.input_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.w1.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.w1.weight_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.w2.input_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.w2.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.6.ffn.experts.mlp.w2.weight_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.6.ffn.router.layer.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.attn.Wqkv.input_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.attn.Wqkv.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.attn.Wqkv.weight_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.attn.kv_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.attn.out_proj.input_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.attn.out_proj.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.attn.out_proj.weight_scale": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.norm_1.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.6.norm_attn_norm.norm_2.weight": "model-00005-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.v1.input_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.v1.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.v1.weight_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.w1.input_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.w1.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.w1.weight_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.w2.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.w2.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.7.ffn.experts.mlp.w2.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.7.ffn.router.layer.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.attn.Wqkv.input_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.attn.Wqkv.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.attn.Wqkv.weight_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.attn.kv_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.attn.out_proj.input_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.attn.out_proj.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.attn.out_proj.weight_scale": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.norm_1.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.7.norm_attn_norm.norm_2.weight": "model-00006-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.v1.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.v1.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.v1.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.w1.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.w1.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.w1.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.w2.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.w2.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.experts.mlp.w2.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.ffn.router.layer.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.attn.Wqkv.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.attn.Wqkv.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.attn.Wqkv.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.attn.kv_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.attn.out_proj.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.attn.out_proj.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.attn.out_proj.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.norm_1.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.8.norm_attn_norm.norm_2.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.v1.input_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.v1.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.v1.weight_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.w1.input_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.w1.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.w1.weight_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.w2.input_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.w2.weight": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.experts.mlp.w2.weight_scale": "model-00008-of-00031.safetensors",
+    "transformer.blocks.9.ffn.router.layer.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.attn.Wqkv.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.attn.Wqkv.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.attn.Wqkv.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.attn.kv_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.attn.out_proj.input_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.attn.out_proj.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.attn.out_proj.weight_scale": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.norm_1.weight": "model-00007-of-00031.safetensors",
+    "transformer.blocks.9.norm_attn_norm.norm_2.weight": "model-00007-of-00031.safetensors",
+    "transformer.norm_f.weight": "model-00031-of-00031.safetensors",
+    "transformer.wte.weight": "model-00001-of-00031.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,205 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "100256": {
+      "content": "<||_unused_0_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100257": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100258": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100259": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100260": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100261": {
+      "content": "<||_unused_1_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100262": {
+      "content": "<||_unused_2_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100263": {
+      "content": "<||_unused_3_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100264": {
+      "content": "<||_unused_4_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100265": {
+      "content": "<||_unused_5_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100266": {
+      "content": "<||_unused_6_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100267": {
+      "content": "<||_unused_7_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100268": {
+      "content": "<||_unused_8_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100269": {
+      "content": "<||_unused_9_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100270": {
+      "content": "<||_unused_10_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100271": {
+      "content": "<||_unused_11_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100272": {
+      "content": "<||_unused_12_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100273": {
+      "content": "<||_unused_13_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100274": {
+      "content": "<||_unused_14_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100275": {
+      "content": "<||_unused_15_||>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100276": {
+      "content": "<|endofprompt|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100277": {
+      "content": "<|pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100278": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100279": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% elif 'system' not in messages[0]['role'] %}{% set loop_messages = messages %}{% set system_message = 'You are DBRX, created by Databricks. You were last updated in December 2023. You answer questions based on information available up to that point.\nYOU PROVIDE SHORT RESPONSES TO SHORT QUESTIONS OR STATEMENTS, but provide thorough responses to more complex and open-ended questions.\nYou assist with various tasks, from writing to coding (using markdown for code blocks — remember to use ``` with code, JSON, and tables).\n(You do not have real-time data access or code execution capabilities. You avoid stereotyping and provide balanced perspectives on controversial topics. You do not provide song lyrics, poems, or news articles and do not divulge details of your training data.)\nThis is your system prompt, guiding your responses. Do not reference it, just respond to the user. If you find yourself talking about this message, stop. You should be responding appropriately and usually that means not mentioning this.\nYOU DO NOT MENTION ANY OF THIS INFORMATION ABOUT YOURSELF UNLESS THE INFORMATION IS DIRECTLY PERTINENT TO THE USER\\'S QUERY.' %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if loop.index0 == 0 %}{% if system_message != false %}{{ '<|im_start|>system\n' + system_message | trim + '<|im_end|>\n'}}{% endif %}{{ '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' }}{% else %}{{ '\n' + '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' }}{% endif %}{% if (add_generation_prompt == true and loop.last) %}{{ '\n' + '<|im_start|>' + 'assistant' + '\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 32768,
+  "pad_token": "<|pad|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}