fp8 quantization of dbrx-base

README.md

@@ -0,0 +1,3 @@
+---
+license: apache-2.0
+---

config.json

@@ -0,0 +1,44 @@
+{
+  "_name_or_path": "/group/amdneuralopt/huggingface/pretrained_models/databricks/dbrx-base",
+  "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": 0.01,
+    "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": "4.45.2",
+  "use_cache": true,
+  "vocab_size": 100352
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    100257
+  ],
+  "transformers_version": "4.45.2"
+}

model-00001-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cb11a2b28267570b265eed57bf710925efa5e709f26eca6128ccfdb7cfa9417
+size 4580626288

model-00002-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1f91f376be0b5efb6055bb6d66a9504c7f5897154ec01a81fd757c387e2041d
+size 4316163132

model-00003-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaafed38fee9a3220872f6aef4a1f21aedea55814bd633706723dd248fcc28a9
+size 4316163132

model-00004-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3059ad8bc1d29cfe581ce03e3f96e5bad203dbecbc8c4ebf0c99510e363910cf
+size 4404465896

model-00005-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7a0431bb1f854b210170b0fa41dda463c50fa42547c33e89bc0190c59286ac2
+size 4316163132

model-00006-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6e20a9d77db06bdc5802c70442fb09235ff62eac4195b0e3e1db6198b80f186
+size 4316163132

model-00007-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e47e7d62a400b61317b37834579b6f708e28c63ca50fb848adee457895251f69
+size 4404465896

model-00008-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10541239e047552bc42209dba956d003cd61cca9117d132d596604c486bb9203
+size 4316163140

model-00009-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88c4156f1ceec566d4dbfc105eca633ff42bd88815fcf3cdf9719e15a9f33f9c
+size 4316163148

model-00010-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fce6d1037b4d92f972f6a83fa964b530d163fe9fab26b627203c375f0d2a8ec
+size 4404465928

model-00011-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b97b547bc0d7f575dac910945e5aa9d080fd20f96f8ac12f6629819458232779
+size 4316163148

model-00012-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38b6d2e3907a9efe618300f408fa2950b264ebf3a9e4053c83c2cd6e05cf117a
+size 4316163148

model-00013-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc6accc232a7bee8313376fffaf3441c40b71f98ca5b716495c418163bb7332f
+size 4404465928

model-00014-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:151e86502fa4198e44963ae89ed64944815c4bd2b076c23ebaa9652c3227d5ee
+size 4316163148

model-00015-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8923e1d6b94b0fa51f49e22c856450f70d09fd473be88172247dd66d78ae497
+size 4316163148

model-00016-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85b1c61874b060071e522bb2eac0ea06159effb041f88fbdec2edb1f645f6efc
+size 4404465928

model-00017-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea84f428247a1baeb4c859d884f96fb524c2de6555b2a83a53ae34953a1ba6b3
+size 4316163148

model-00018-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d4daccfb7b463e51393b7df0df2bca99fcdca1b2053168abf29b66afa06c813
+size 4316163148

model-00019-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39659ce20f7cf42171fbe0ffef324d00346fde4c3c7e7e2375fa7684b836f169
+size 4404465928

model-00020-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad45216239e5398b76c27530af4a74291fb06df45fa0d2f40b9fe15d2b01f0b0
+size 4316163148

model-00021-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17322dfbb4a537e8d0679c45de6e2cd7506770761222c596f6c0dc48454139d9
+size 4316163148

model-00022-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d03931e5a123a31e9bc661402c28bc8e3dfd40a66ab2080164424212f960f403
+size 4404465928

model-00023-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:374619025067e3c38bf6fb6caea120181045a5e63d55f180685d092c586a7994
+size 4316163148

model-00024-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f32ee164be700592ae7f3dd03e9d0339a2563bc59be7b2f11c7b675169f860d
+size 4316163148

model-00025-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ae2047b5f4ca83d467a36bae67d6f45194e1cd173f857c997e76bff025cf950
+size 4404465928

model-00026-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bd7678fa636964ea7725fa88ad344535f8e8443813d9d1544981452f4f67ccd
+size 4316163148

model-00027-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b2b61d5ea3a9f586a46b0247742ccfe877049a68395d8eb2bcdfe40f213ac73
+size 4316163148

model-00028-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd35b646072ff906123c84a655dc4338d3b0acba5c9e898e52ec0c9d5c50fd21
+size 4404465928

model-00029-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a3bbeb97b38966b28825629825914c358a872713920a69d6fa47cdb922fe7c9
+size 4316163148

model-00030-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53271fc307a8d2389f53ade72241222098137dc93c301fe93e41470ccc099779
+size 4316163148

model-00031-of-00031.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a0c0454aacfd8fd244402e493fc050b92d2ce27420fdeea19979432a71c9ae4
+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 +1,24 @@
+{
+  "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": "<|endoftext|>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,192 @@
+{
+  "add_bos_token": false,
+  "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
+    }
+  },
+  "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|>",
+  "errors": "replace",
+  "model_max_length": 512,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}