Upload 38 files

config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "TomGrc/FusionNet_34Bx2_MoE",
+  "architectures": [
+    "MixtralForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 7168,
+  "initializer_range": 0.02,
+  "intermediate_size": 20480,
+  "max_position_embeddings": 32768,
+  "model_type": "mixtral",
+  "num_attention_heads": 56,
+  "num_experts_per_tok": 2,
+  "num_hidden_layers": 60,
+  "num_key_value_heads": 8,
+  "num_local_experts": 2,
+  "output_router_logits": false,
+  "pad_token_id": 1,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 5000000.0,
+  "router_aux_loss_coef": 0.001,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.36.2",
+  "use_cache": true,
+  "vocab_size": 64000
+}

model-00001-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2fbed1fc3ecc13e59701eca6f6fcd4f90ec0ed68e4a12f07489d40cbe90cded
+size 3891099408

model-00002-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e97dd85c18a92741a1949a42f21ee559b3df3bb6eda7017ff4e8481aad90d519
+size 3816818368

model-00003-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1831f264cbec5212a5fb2725ef476c066ec95fdf9ef92570984785171d519c63
+size 3816818384

model-00004-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93369b670a5c1c1da81aeba09423e33086b60cef9176a04881b197b95a18e682
+size 3816818384

model-00005-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a2fbe62884df91749225e84c01e56e643bd76891a1408a1c3dbee35c466e34f
+size 3816818384

model-00006-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3cd2fa4ee4c9d0d1b0569f016be0ba6c2dc78c6bdca34c377faa35b6694ea52
+size 3816818384

model-00007-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea879cd1cfb68c77d1cb53d61ac128ee47cd17a42866909cd6d4dc06317dd669
+size 3816818384

model-00008-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a95dcaa2e8a9c46c373b86e8011099b513a31a4357342074aba5d9b8b652acc
+size 3816818384

model-00009-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bfbf228bf7636d7a6264aa503722843d64c2c8f7326dd6a8d078c967a07da36
+size 3816818384

model-00010-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4731f22c0c7f52947d112ceabb46e08beb7dfc2eeb04f0e530641b282480c249
+size 3816818384

model-00011-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6980e0fd16f09da7fa8ca67ff194897fc0c52293165fbdbdfb2719bbb28ab358
+size 3816818368

model-00012-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd86a7984ff3bc7b820ea1992bfffc8af568c891e186950f51f663f87144e667
+size 3816818384

model-00013-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d489b276b5aae6b917f63c8560009db138850c7710841c6991798a9484f510a
+size 3816818384

model-00014-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f4afb6cd29a7f4fee671d69dfc4bbaed8e6f7e14a4e27238c7cbe12c00c3cc3
+size 3816818384

model-00015-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9df44a85e711e93af3627db76f846027f00a486b21abe64d90efdd73d47cdce
+size 3816818384

model-00016-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46f6e29fc7f66dfbafe0cecea3c41cdece96c239565dd8bf070258a93f3706ea
+size 3816818384

model-00017-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b07e70485691d20b1838e282553fafe51026829b295536855cdaca06cd5e5610
+size 3816818384

model-00018-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:493230329db95949525d9d9d2114dfad0a61cda4951d1486e350008b0448d2f8
+size 3816818384

model-00019-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34057d877023fcbac6ac9a7e81685b6b674e87648d3cf08b5eb553897b95b927
+size 3816818384

model-00020-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eab5eb8cf885d55d4213ee8ac38a5353b0aa7bf6f338c350a87914c734e90199
+size 3816818368

model-00021-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51b2461b041a0a8a2800261b125297bf80592f3b65e548c347f31963860e8684
+size 3816818376

model-00022-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89cec10e2ce8312d5bfec77b303af9ebf59d00035ab14f2899f17f0ddef9ebf2
+size 3816818384

model-00023-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:650660cb28b07ababdc24580cf6e7c27dd7611039070d75e94acb502be91f4f4
+size 3816818384

model-00024-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca0ece90cdcc407f5166a62e50e14c3177209ef9e0542e0b5baddfbac3bedd66
+size 3816818384

model-00025-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45e6df2569d5d4e1ca04367f23fbe43d9495515f6ce36c1863482aece7e148a7
+size 3816818384

model-00026-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86db85673d0547eb5e1711c5685574f32f45d02961d256aab578d66b0627491a
+size 3816818384

model-00027-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:456a8e50274f11b0bf78ab1e7e0b6cdd056309c0a9f50434f898bd587a0ca86e
+size 3816818384

model-00028-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e4aa4c15b57b4a02a99b078327fe992769246ca066805b8740685afcf32f4c2
+size 3816818384

model-00029-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82f182510c5331bcf0c37deeade85d2a0cc92dd0f1321dde05273975eb7c6ce7
+size 3979168888

model-00030-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2128bf49875a054515a7d3e8ae9459127ed57655afd60a00db6878a91426ba7d
+size 3992990288

model-00031-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a4a08c979860345aae1db03199d04cc3d6f7c08c32587545fc7a7e04b20bfd9
+size 3934265416

model-00032-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a07b7cef78260ecf6b2b26b83a9d985d1e2a28c077f14caee56ae7ba1fda50f
+size 2776262560

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.3.2"}, "weight_map": {"model.embed_tokens.weight": "model-00001-of-00032.safetensors", "model.norm.weight": "model-00001-of-00032.safetensors", "lm_head.weight": "model-00001-of-00032.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.5.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.6.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.7.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.8.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.9.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.11.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.12.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.13.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.14.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.15.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.16.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.17.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.18.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.19.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.20.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.21.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.22.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.23.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.24.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.25.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.26.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.27.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.28.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.29.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.30.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.31.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.32.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.33.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.34.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.35.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.36.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.37.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.38.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.39.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.40.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.41.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.42.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.43.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.44.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.45.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.46.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.47.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.48.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.49.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.50.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.51.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.52.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.53.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.54.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.55.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.56.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.57.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.58.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.59.input_layernorm.weight": "model-00001-of-00032.safetensors", "model.layers.0.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00032.safetensors", "model.layers.0.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00032.safetensors", "model.layers.1.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00032.safetensors", "model.layers.1.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00032.safetensors", "model.layers.2.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00032.safetensors", "model.layers.2.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00032.safetensors", "model.layers.3.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00032.safetensors", "model.layers.3.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00032.safetensors", "model.layers.4.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00032.safetensors", "model.layers.4.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00032.safetensors", "model.layers.5.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00032.safetensors", "model.layers.5.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00032.safetensors", "model.layers.6.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00032.safetensors", "model.layers.6.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00032.safetensors", "model.layers.7.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00032.safetensors", "model.layers.7.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00032.safetensors", "model.layers.8.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00032.safetensors", "model.layers.8.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00032.safetensors", "model.layers.9.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00032.safetensors", "model.layers.9.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00032.safetensors", "model.layers.10.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00032.safetensors", "model.layers.10.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00032.safetensors", "model.layers.11.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00032.safetensors", "model.layers.11.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00032.safetensors", "model.layers.12.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00032.safetensors", "model.layers.12.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00032.safetensors", "model.layers.13.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00032.safetensors", "model.layers.13.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00032.safetensors", "model.layers.14.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00032.safetensors", "model.layers.14.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00032.safetensors", "model.layers.15.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00032.safetensors", "model.layers.15.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00032.safetensors", "model.layers.16.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00032.safetensors", "model.layers.16.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00032.safetensors", "model.layers.17.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00032.safetensors", "model.layers.17.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00032.safetensors", "model.layers.18.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00032.safetensors", "model.layers.18.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00032.safetensors", "model.layers.19.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00032.safetensors", "model.layers.19.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00032.safetensors", "model.layers.20.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00032.safetensors", "model.layers.20.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00032.safetensors", "model.layers.21.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00032.safetensors", "model.layers.21.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00032.safetensors", "model.layers.22.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00032.safetensors", "model.layers.22.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00032.safetensors", "model.layers.23.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00032.safetensors", "model.layers.23.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00032.safetensors", "model.layers.24.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00032.safetensors", "model.layers.24.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00032.safetensors", "model.layers.25.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00032.safetensors", "model.layers.25.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00032.safetensors", "model.layers.26.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00032.safetensors", "model.layers.26.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00032.safetensors", "model.layers.27.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00032.safetensors", "model.layers.27.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00032.safetensors", "model.layers.28.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00032.safetensors", "model.layers.28.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00032.safetensors", "model.layers.29.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00032.safetensors", "model.layers.29.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00032.safetensors", "model.layers.30.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00032.safetensors", "model.layers.30.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00032.safetensors", "model.layers.31.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00032.safetensors", "model.layers.31.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00032.safetensors", "model.layers.32.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00032.safetensors", "model.layers.32.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00032.safetensors", "model.layers.33.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00032.safetensors", "model.layers.33.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00032.safetensors", "model.layers.34.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00032.safetensors", "model.layers.34.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00032.safetensors", "model.layers.35.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00032.safetensors", "model.layers.35.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00032.safetensors", "model.layers.36.block_sparse_moe.experts.0.w3.weight": "model-00007-of-00032.safetensors", "model.layers.36.block_sparse_moe.experts.1.w3.weight": "model-00007-of-00032.safetensors", "model.layers.37.block_sparse_moe.experts.0.w3.weight": "model-00007-of-00032.safetensors", "model.layers.37.block_sparse_moe.experts.1.w3.weight": "model-00007-of-00032.safetensors", "model.layers.38.block_sparse_moe.experts.0.w3.weight": "model-00007-of-00032.safetensors", "model.layers.38.block_sparse_moe.experts.1.w3.weight": "model-00007-of-00032.safetensors", "model.layers.39.block_sparse_moe.experts.0.w3.weight": "model-00007-of-00032.safetensors", "model.layers.39.block_sparse_moe.experts.1.w3.weight": "model-00007-of-00032.safetensors", "model.layers.40.block_sparse_moe.experts.0.w3.weight": "model-00007-of-00032.safetensors", "model.layers.40.block_sparse_moe.experts.1.w3.weight": "model-00007-of-00032.safetensors", "model.layers.41.block_sparse_moe.experts.0.w3.weight": "model-00007-of-00032.safetensors", "model.layers.41.block_sparse_moe.experts.1.w3.weight": "model-00007-of-00032.safetensors", "model.layers.42.block_sparse_moe.experts.0.w3.weight": "model-00007-of-00032.safetensors", "model.layers.42.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00032.safetensors", "model.layers.43.block_sparse_moe.experts.0.w3.weight": "model-00008-of-00032.safetensors", "model.layers.43.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00032.safetensors", "model.layers.44.block_sparse_moe.experts.0.w3.weight": "model-00008-of-00032.safetensors", "model.layers.44.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00032.safetensors", "model.layers.45.block_sparse_moe.experts.0.w3.weight": "model-00008-of-00032.safetensors", "model.layers.45.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00032.safetensors", "model.layers.46.block_sparse_moe.experts.0.w3.weight": "model-00008-of-00032.safetensors", "model.layers.46.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00032.safetensors", "model.layers.47.block_sparse_moe.experts.0.w3.weight": "model-00008-of-00032.safetensors", "model.layers.47.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00032.safetensors", "model.layers.48.block_sparse_moe.experts.0.w3.weight": "model-00008-of-00032.safetensors", "model.layers.48.block_sparse_moe.experts.1.w3.weight": "model-00008-of-00032.safetensors", "model.layers.49.block_sparse_moe.experts.0.w3.weight": "model-00009-of-00032.safetensors", "model.layers.49.block_sparse_moe.experts.1.w3.weight": "model-00009-of-00032.safetensors", "model.layers.50.block_sparse_moe.experts.0.w3.weight": "model-00009-of-00032.safetensors", "model.layers.50.block_sparse_moe.experts.1.w3.weight": "model-00009-of-00032.safetensors", "model.layers.51.block_sparse_moe.experts.0.w3.weight": "model-00009-of-00032.safetensors", "model.layers.51.block_sparse_moe.experts.1.w3.weight": "model-00009-of-00032.safetensors", "model.layers.52.block_sparse_moe.experts.0.w3.weight": "model-00009-of-00032.safetensors", "model.layers.52.block_sparse_moe.experts.1.w3.weight": "model-00009-of-00032.safetensors", "model.layers.53.block_sparse_moe.experts.0.w3.weight": "model-00009-of-00032.safetensors", "model.layers.53.block_sparse_moe.experts.1.w3.weight": "model-00009-of-00032.safetensors", "model.layers.54.block_sparse_moe.experts.0.w3.weight": "model-00009-of-00032.safetensors", "model.layers.54.block_sparse_moe.experts.1.w3.weight": "model-00009-of-00032.safetensors", "model.layers.55.block_sparse_moe.experts.0.w3.weight": "model-00009-of-00032.safetensors", "model.layers.55.block_sparse_moe.experts.1.w3.weight": "model-00010-of-00032.safetensors", "model.layers.56.block_sparse_moe.experts.0.w3.weight": "model-00010-of-00032.safetensors", "model.layers.56.block_sparse_moe.experts.1.w3.weight": "model-00010-of-00032.safetensors", "model.layers.57.block_sparse_moe.experts.0.w3.weight": "model-00010-of-00032.safetensors", "model.layers.57.block_sparse_moe.experts.1.w3.weight": "model-00010-of-00032.safetensors", "model.layers.58.block_sparse_moe.experts.0.w3.weight": "model-00010-of-00032.safetensors", "model.layers.58.block_sparse_moe.experts.1.w3.weight": "model-00010-of-00032.safetensors", "model.layers.59.block_sparse_moe.experts.0.w3.weight": "model-00010-of-00032.safetensors", "model.layers.59.block_sparse_moe.experts.1.w3.weight": "model-00010-of-00032.safetensors", "model.layers.0.block_sparse_moe.experts.0.w2.weight": "model-00010-of-00032.safetensors", "model.layers.0.block_sparse_moe.experts.1.w2.weight": "model-00010-of-00032.safetensors", "model.layers.1.block_sparse_moe.experts.0.w2.weight": "model-00010-of-00032.safetensors", "model.layers.1.block_sparse_moe.experts.1.w2.weight": "model-00010-of-00032.safetensors", "model.layers.2.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00032.safetensors", "model.layers.2.block_sparse_moe.experts.1.w2.weight": "model-00011-of-00032.safetensors", "model.layers.3.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00032.safetensors", "model.layers.3.block_sparse_moe.experts.1.w2.weight": "model-00011-of-00032.safetensors", "model.layers.4.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00032.safetensors", "model.layers.4.block_sparse_moe.experts.1.w2.weight": "model-00011-of-00032.safetensors", "model.layers.5.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00032.safetensors", "model.layers.5.block_sparse_moe.experts.1.w2.weight": "model-00011-of-00032.safetensors", "model.layers.6.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00032.safetensors", "model.layers.6.block_sparse_moe.experts.1.w2.weight": "model-00011-of-00032.safetensors", "model.layers.7.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00032.safetensors", "model.layers.7.block_sparse_moe.experts.1.w2.weight": "model-00011-of-00032.safetensors", "model.layers.8.block_sparse_moe.experts.0.w2.weight": "model-00011-of-00032.safetensors", "model.layers.8.block_sparse_moe.experts.1.w2.weight": "model-00012-of-00032.safetensors", "model.layers.9.block_sparse_moe.experts.0.w2.weight": "model-00012-of-00032.safetensors", "model.layers.9.block_sparse_moe.experts.1.w2.weight": "model-00012-of-00032.safetensors", "model.layers.10.block_sparse_moe.experts.0.w2.weight": "model-00012-of-00032.safetensors", "model.layers.10.block_sparse_moe.experts.1.w2.weight": "model-00012-of-00032.safetensors", "model.layers.11.block_sparse_moe.experts.0.w2.weight": "model-00012-of-00032.safetensors", "model.layers.11.block_sparse_moe.experts.1.w2.weight": "model-00012-of-00032.safetensors", "model.layers.12.block_sparse_moe.experts.0.w2.weight": "model-00012-of-00032.safetensors", "model.layers.12.block_sparse_moe.experts.1.w2.weight": "model-00012-of-00032.safetensors", "model.layers.13.block_sparse_moe.experts.0.w2.weight": "model-00012-of-00032.safetensors", "model.layers.13.block_sparse_moe.experts.1.w2.weight": "model-00012-of-00032.safetensors", "model.layers.14.block_sparse_moe.experts.0.w2.weight": "model-00012-of-00032.safetensors", "model.layers.14.block_sparse_moe.experts.1.w2.weight": "model-00012-of-00032.safetensors", "model.layers.15.block_sparse_moe.experts.0.w2.weight": "model-00013-of-00032.safetensors", "model.layers.15.block_sparse_moe.experts.1.w2.weight": "model-00013-of-00032.safetensors", "model.layers.16.block_sparse_moe.experts.0.w2.weight": "model-00013-of-00032.safetensors", "model.layers.16.block_sparse_moe.experts.1.w2.weight": "model-00013-of-00032.safetensors", "model.layers.17.block_sparse_moe.experts.0.w2.weight": "model-00013-of-00032.safetensors", "model.layers.17.block_sparse_moe.experts.1.w2.weight": "model-00013-of-00032.safetensors", "model.layers.18.block_sparse_moe.experts.0.w2.weight": "model-00013-of-00032.safetensors", "model.layers.18.block_sparse_moe.experts.1.w2.weight": "model-00013-of-00032.safetensors", "model.layers.19.block_sparse_moe.experts.0.w2.weight": "model-00013-of-00032.safetensors", "model.layers.19.block_sparse_moe.experts.1.w2.weight": "model-00013-of-00032.safetensors", "model.layers.20.block_sparse_moe.experts.0.w2.weight": "model-00013-of-00032.safetensors", "model.layers.20.block_sparse_moe.experts.1.w2.weight": "model-00013-of-00032.safetensors", "model.layers.21.block_sparse_moe.experts.0.w2.weight": "model-00013-of-00032.safetensors", "model.layers.21.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00032.safetensors", "model.layers.22.block_sparse_moe.experts.0.w2.weight": "model-00014-of-00032.safetensors", "model.layers.22.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00032.safetensors", "model.layers.23.block_sparse_moe.experts.0.w2.weight": "model-00014-of-00032.safetensors", "model.layers.23.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00032.safetensors", "model.layers.24.block_sparse_moe.experts.0.w2.weight": "model-00014-of-00032.safetensors", "model.layers.24.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00032.safetensors", "model.layers.25.block_sparse_moe.experts.0.w2.weight": "model-00014-of-00032.safetensors", "model.layers.25.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00032.safetensors", "model.layers.26.block_sparse_moe.experts.0.w2.weight": "model-00014-of-00032.safetensors", "model.layers.26.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00032.safetensors", "model.layers.27.block_sparse_moe.experts.0.w2.weight": "model-00014-of-00032.safetensors", "model.layers.27.block_sparse_moe.experts.1.w2.weight": "model-00014-of-00032.safetensors", "model.layers.28.block_sparse_moe.experts.0.w2.weight": "model-00015-of-00032.safetensors", "model.layers.28.block_sparse_moe.experts.1.w2.weight": "model-00015-of-00032.safetensors", "model.layers.29.block_sparse_moe.experts.0.w2.weight": "model-00015-of-00032.safetensors", "model.layers.29.block_sparse_moe.experts.1.w2.weight": "model-00015-of-00032.safetensors", "model.layers.30.block_sparse_moe.experts.0.w2.weight": "model-00015-of-00032.safetensors", "model.layers.30.block_sparse_moe.experts.1.w2.weight": "model-00015-of-00032.safetensors", "model.layers.31.block_sparse_moe.experts.0.w2.weight": "model-00015-of-00032.safetensors", "model.layers.31.block_sparse_moe.experts.1.w2.weight": "model-00015-of-00032.safetensors", "model.layers.32.block_sparse_moe.experts.0.w2.weight": "model-00015-of-00032.safetensors", "model.layers.32.block_sparse_moe.experts.1.w2.weight": "model-00015-of-00032.safetensors", "model.layers.33.block_sparse_moe.experts.0.w2.weight": "model-00015-of-00032.safetensors", "model.layers.33.block_sparse_moe.experts.1.w2.weight": "model-00015-of-00032.safetensors", "model.layers.34.block_sparse_moe.experts.0.w2.weight": "model-00015-of-00032.safetensors", "model.layers.34.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00032.safetensors", "model.layers.35.block_sparse_moe.experts.0.w2.weight": "model-00016-of-00032.safetensors", "model.layers.35.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00032.safetensors", "model.layers.36.block_sparse_moe.experts.0.w2.weight": "model-00016-of-00032.safetensors", "model.layers.36.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00032.safetensors", "model.layers.37.block_sparse_moe.experts.0.w2.weight": "model-00016-of-00032.safetensors", "model.layers.37.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00032.safetensors", "model.layers.38.block_sparse_moe.experts.0.w2.weight": "model-00016-of-00032.safetensors", "model.layers.38.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00032.safetensors", "model.layers.39.block_sparse_moe.experts.0.w2.weight": "model-00016-of-00032.safetensors", "model.layers.39.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00032.safetensors", "model.layers.40.block_sparse_moe.experts.0.w2.weight": "model-00016-of-00032.safetensors", "model.layers.40.block_sparse_moe.experts.1.w2.weight": "model-00016-of-00032.safetensors", "model.layers.41.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00032.safetensors", "model.layers.41.block_sparse_moe.experts.1.w2.weight": "model-00017-of-00032.safetensors", "model.layers.42.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00032.safetensors", "model.layers.42.block_sparse_moe.experts.1.w2.weight": "model-00017-of-00032.safetensors", "model.layers.43.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00032.safetensors", "model.layers.43.block_sparse_moe.experts.1.w2.weight": "model-00017-of-00032.safetensors", "model.layers.44.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00032.safetensors", "model.layers.44.block_sparse_moe.experts.1.w2.weight": "model-00017-of-00032.safetensors", "model.layers.45.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00032.safetensors", "model.layers.45.block_sparse_moe.experts.1.w2.weight": "model-00017-of-00032.safetensors", "model.layers.46.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00032.safetensors", "model.layers.46.block_sparse_moe.experts.1.w2.weight": "model-00017-of-00032.safetensors", "model.layers.47.block_sparse_moe.experts.0.w2.weight": "model-00017-of-00032.safetensors", "model.layers.47.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00032.safetensors", "model.layers.48.block_sparse_moe.experts.0.w2.weight": "model-00018-of-00032.safetensors", "model.layers.48.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00032.safetensors", "model.layers.49.block_sparse_moe.experts.0.w2.weight": "model-00018-of-00032.safetensors", "model.layers.49.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00032.safetensors", "model.layers.50.block_sparse_moe.experts.0.w2.weight": "model-00018-of-00032.safetensors", "model.layers.50.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00032.safetensors", "model.layers.51.block_sparse_moe.experts.0.w2.weight": "model-00018-of-00032.safetensors", "model.layers.51.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00032.safetensors", "model.layers.52.block_sparse_moe.experts.0.w2.weight": "model-00018-of-00032.safetensors", "model.layers.52.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00032.safetensors", "model.layers.53.block_sparse_moe.experts.0.w2.weight": "model-00018-of-00032.safetensors", "model.layers.53.block_sparse_moe.experts.1.w2.weight": "model-00018-of-00032.safetensors", "model.layers.54.block_sparse_moe.experts.0.w2.weight": "model-00019-of-00032.safetensors", "model.layers.54.block_sparse_moe.experts.1.w2.weight": "model-00019-of-00032.safetensors", "model.layers.55.block_sparse_moe.experts.0.w2.weight": "model-00019-of-00032.safetensors", "model.layers.55.block_sparse_moe.experts.1.w2.weight": "model-00019-of-00032.safetensors", "model.layers.56.block_sparse_moe.experts.0.w2.weight": "model-00019-of-00032.safetensors", "model.layers.56.block_sparse_moe.experts.1.w2.weight": "model-00019-of-00032.safetensors", "model.layers.57.block_sparse_moe.experts.0.w2.weight": "model-00019-of-00032.safetensors", "model.layers.57.block_sparse_moe.experts.1.w2.weight": "model-00019-of-00032.safetensors", "model.layers.58.block_sparse_moe.experts.0.w2.weight": "model-00019-of-00032.safetensors", "model.layers.58.block_sparse_moe.experts.1.w2.weight": "model-00019-of-00032.safetensors", "model.layers.59.block_sparse_moe.experts.0.w2.weight": "model-00019-of-00032.safetensors", "model.layers.59.block_sparse_moe.experts.1.w2.weight": "model-00019-of-00032.safetensors", "model.layers.0.block_sparse_moe.experts.0.w1.weight": "model-00019-of-00032.safetensors", "model.layers.0.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00032.safetensors", "model.layers.1.block_sparse_moe.experts.0.w1.weight": "model-00020-of-00032.safetensors", "model.layers.1.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00032.safetensors", "model.layers.2.block_sparse_moe.experts.0.w1.weight": "model-00020-of-00032.safetensors", "model.layers.2.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00032.safetensors", "model.layers.3.block_sparse_moe.experts.0.w1.weight": "model-00020-of-00032.safetensors", "model.layers.3.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00032.safetensors", "model.layers.4.block_sparse_moe.experts.0.w1.weight": "model-00020-of-00032.safetensors", "model.layers.4.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00032.safetensors", "model.layers.5.block_sparse_moe.experts.0.w1.weight": "model-00020-of-00032.safetensors", "model.layers.5.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00032.safetensors", "model.layers.6.block_sparse_moe.experts.0.w1.weight": "model-00020-of-00032.safetensors", "model.layers.6.block_sparse_moe.experts.1.w1.weight": "model-00020-of-00032.safetensors", "model.layers.7.block_sparse_moe.experts.0.w1.weight": "model-00021-of-00032.safetensors", "model.layers.7.block_sparse_moe.experts.1.w1.weight": "model-00021-of-00032.safetensors", "model.layers.8.block_sparse_moe.experts.0.w1.weight": "model-00021-of-00032.safetensors", "model.layers.8.block_sparse_moe.experts.1.w1.weight": "model-00021-of-00032.safetensors", "model.layers.9.block_sparse_moe.experts.0.w1.weight": "model-00021-of-00032.safetensors", "model.layers.9.block_sparse_moe.experts.1.w1.weight": "model-00021-of-00032.safetensors", "model.layers.10.block_sparse_moe.experts.0.w1.weight": "model-00021-of-00032.safetensors", "model.layers.10.block_sparse_moe.experts.1.w1.weight": "model-00021-of-00032.safetensors", "model.layers.11.block_sparse_moe.experts.0.w1.weight": "model-00021-of-00032.safetensors", "model.layers.11.block_sparse_moe.experts.1.w1.weight": "model-00021-of-00032.safetensors", "model.layers.12.block_sparse_moe.experts.0.w1.weight": "model-00021-of-00032.safetensors", "model.layers.12.block_sparse_moe.experts.1.w1.weight": "model-00021-of-00032.safetensors", "model.layers.13.block_sparse_moe.experts.0.w1.weight": "model-00021-of-00032.safetensors", "model.layers.13.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00032.safetensors", "model.layers.14.block_sparse_moe.experts.0.w1.weight": "model-00022-of-00032.safetensors", "model.layers.14.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00032.safetensors", "model.layers.15.block_sparse_moe.experts.0.w1.weight": "model-00022-of-00032.safetensors", "model.layers.15.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00032.safetensors", "model.layers.16.block_sparse_moe.experts.0.w1.weight": "model-00022-of-00032.safetensors", "model.layers.16.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00032.safetensors", "model.layers.17.block_sparse_moe.experts.0.w1.weight": "model-00022-of-00032.safetensors", "model.layers.17.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00032.safetensors", "model.layers.18.block_sparse_moe.experts.0.w1.weight": "model-00022-of-00032.safetensors", "model.layers.18.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00032.safetensors", "model.layers.19.block_sparse_moe.experts.0.w1.weight": "model-00022-of-00032.safetensors", "model.layers.19.block_sparse_moe.experts.1.w1.weight": "model-00022-of-00032.safetensors", "model.layers.20.block_sparse_moe.experts.0.w1.weight": "model-00023-of-00032.safetensors", "model.layers.20.block_sparse_moe.experts.1.w1.weight": "model-00023-of-00032.safetensors", "model.layers.21.block_sparse_moe.experts.0.w1.weight": "model-00023-of-00032.safetensors", "model.layers.21.block_sparse_moe.experts.1.w1.weight": "model-00023-of-00032.safetensors", "model.layers.22.block_sparse_moe.experts.0.w1.weight": "model-00023-of-00032.safetensors", "model.layers.22.block_sparse_moe.experts.1.w1.weight": "model-00023-of-00032.safetensors", "model.layers.23.block_sparse_moe.experts.0.w1.weight": "model-00023-of-00032.safetensors", "model.layers.23.block_sparse_moe.experts.1.w1.weight": "model-00023-of-00032.safetensors", "model.layers.24.block_sparse_moe.experts.0.w1.weight": "model-00023-of-00032.safetensors", "model.layers.24.block_sparse_moe.experts.1.w1.weight": "model-00023-of-00032.safetensors", "model.layers.25.block_sparse_moe.experts.0.w1.weight": "model-00023-of-00032.safetensors", "model.layers.25.block_sparse_moe.experts.1.w1.weight": "model-00023-of-00032.safetensors", "model.layers.26.block_sparse_moe.experts.0.w1.weight": "model-00023-of-00032.safetensors", "model.layers.26.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00032.safetensors", "model.layers.27.block_sparse_moe.experts.0.w1.weight": "model-00024-of-00032.safetensors", "model.layers.27.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00032.safetensors", "model.layers.28.block_sparse_moe.experts.0.w1.weight": "model-00024-of-00032.safetensors", "model.layers.28.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00032.safetensors", "model.layers.29.block_sparse_moe.experts.0.w1.weight": "model-00024-of-00032.safetensors", "model.layers.29.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00032.safetensors", "model.layers.30.block_sparse_moe.experts.0.w1.weight": "model-00024-of-00032.safetensors", "model.layers.30.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00032.safetensors", "model.layers.31.block_sparse_moe.experts.0.w1.weight": "model-00024-of-00032.safetensors", "model.layers.31.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00032.safetensors", "model.layers.32.block_sparse_moe.experts.0.w1.weight": "model-00024-of-00032.safetensors", "model.layers.32.block_sparse_moe.experts.1.w1.weight": "model-00024-of-00032.safetensors", "model.layers.33.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00032.safetensors", "model.layers.33.block_sparse_moe.experts.1.w1.weight": "model-00025-of-00032.safetensors", "model.layers.34.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00032.safetensors", "model.layers.34.block_sparse_moe.experts.1.w1.weight": "model-00025-of-00032.safetensors", "model.layers.35.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00032.safetensors", "model.layers.35.block_sparse_moe.experts.1.w1.weight": "model-00025-of-00032.safetensors", "model.layers.36.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00032.safetensors", "model.layers.36.block_sparse_moe.experts.1.w1.weight": "model-00025-of-00032.safetensors", "model.layers.37.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00032.safetensors", "model.layers.37.block_sparse_moe.experts.1.w1.weight": "model-00025-of-00032.safetensors", "model.layers.38.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00032.safetensors", "model.layers.38.block_sparse_moe.experts.1.w1.weight": "model-00025-of-00032.safetensors", "model.layers.39.block_sparse_moe.experts.0.w1.weight": "model-00025-of-00032.safetensors", "model.layers.39.block_sparse_moe.experts.1.w1.weight": "model-00026-of-00032.safetensors", "model.layers.40.block_sparse_moe.experts.0.w1.weight": "model-00026-of-00032.safetensors", "model.layers.40.block_sparse_moe.experts.1.w1.weight": "model-00026-of-00032.safetensors", "model.layers.41.block_sparse_moe.experts.0.w1.weight": "model-00026-of-00032.safetensors", "model.layers.41.block_sparse_moe.experts.1.w1.weight": "model-00026-of-00032.safetensors", "model.layers.42.block_sparse_moe.experts.0.w1.weight": "model-00026-of-00032.safetensors", "model.layers.42.block_sparse_moe.experts.1.w1.weight": "model-00026-of-00032.safetensors", "model.layers.43.block_sparse_moe.experts.0.w1.weight": "model-00026-of-00032.safetensors", "model.layers.43.block_sparse_moe.experts.1.w1.weight": "model-00026-of-00032.safetensors", "model.layers.44.block_sparse_moe.experts.0.w1.weight": "model-00026-of-00032.safetensors", "model.layers.44.block_sparse_moe.experts.1.w1.weight": "model-00026-of-00032.safetensors", "model.layers.45.block_sparse_moe.experts.0.w1.weight": "model-00026-of-00032.safetensors", "model.layers.45.block_sparse_moe.experts.1.w1.weight": "model-00026-of-00032.safetensors", "model.layers.46.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00032.safetensors", "model.layers.46.block_sparse_moe.experts.1.w1.weight": "model-00027-of-00032.safetensors", "model.layers.47.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00032.safetensors", "model.layers.47.block_sparse_moe.experts.1.w1.weight": "model-00027-of-00032.safetensors", "model.layers.48.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00032.safetensors", "model.layers.48.block_sparse_moe.experts.1.w1.weight": "model-00027-of-00032.safetensors", "model.layers.49.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00032.safetensors", "model.layers.49.block_sparse_moe.experts.1.w1.weight": "model-00027-of-00032.safetensors", "model.layers.50.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00032.safetensors", "model.layers.50.block_sparse_moe.experts.1.w1.weight": "model-00027-of-00032.safetensors", "model.layers.51.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00032.safetensors", "model.layers.51.block_sparse_moe.experts.1.w1.weight": "model-00027-of-00032.safetensors", "model.layers.52.block_sparse_moe.experts.0.w1.weight": "model-00027-of-00032.safetensors", "model.layers.52.block_sparse_moe.experts.1.w1.weight": "model-00028-of-00032.safetensors", "model.layers.53.block_sparse_moe.experts.0.w1.weight": "model-00028-of-00032.safetensors", "model.layers.53.block_sparse_moe.experts.1.w1.weight": "model-00028-of-00032.safetensors", "model.layers.54.block_sparse_moe.experts.0.w1.weight": "model-00028-of-00032.safetensors", "model.layers.54.block_sparse_moe.experts.1.w1.weight": "model-00028-of-00032.safetensors", "model.layers.55.block_sparse_moe.experts.0.w1.weight": "model-00028-of-00032.safetensors", "model.layers.55.block_sparse_moe.experts.1.w1.weight": "model-00028-of-00032.safetensors", "model.layers.56.block_sparse_moe.experts.0.w1.weight": "model-00028-of-00032.safetensors", "model.layers.56.block_sparse_moe.experts.1.w1.weight": "model-00028-of-00032.safetensors", "model.layers.57.block_sparse_moe.experts.0.w1.weight": "model-00028-of-00032.safetensors", "model.layers.57.block_sparse_moe.experts.1.w1.weight": "model-00028-of-00032.safetensors", "model.layers.58.block_sparse_moe.experts.0.w1.weight": "model-00028-of-00032.safetensors", "model.layers.58.block_sparse_moe.experts.1.w1.weight": "model-00028-of-00032.safetensors", "model.layers.59.block_sparse_moe.experts.0.w1.weight": "model-00029-of-00032.safetensors", "model.layers.59.block_sparse_moe.experts.1.w1.weight": "model-00029-of-00032.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00029-of-00032.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00029-of-00032.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00030-of-00032.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00030-of-00032.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00030-of-00032.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00031-of-00032.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00031-of-00032.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00032-of-00032.safetensors", "model.layers.0.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.1.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.2.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.3.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.4.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.5.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.6.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.7.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.8.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.9.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.10.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.11.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.12.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.13.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.14.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.15.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.16.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.17.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.18.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.19.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.20.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.21.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.22.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.23.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.24.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.25.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.26.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.27.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.28.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.29.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.30.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.31.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.32.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.33.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.34.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.35.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.36.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.37.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.38.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.39.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.40.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.41.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.42.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.43.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.44.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.45.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.46.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.47.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.48.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.49.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.50.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.51.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.52.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.53.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.54.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.55.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.56.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.57.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.58.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors", "model.layers.59.block_sparse_moe.gate.weight": "model-00032-of-00032.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:386c49cf943d71aa110361135338c50e38beeff0a66593480421f37b319e1a39
+size 1033105

tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{%- for idx in range(0, messages|length) -%}\n{%- if messages[idx]['role'] == 'user' -%}\n{%- if idx > 1 -%}\n{{- bos_token + '[INST] ' + messages[idx]['content'] + ' [/INST]' -}}\n{%- else -%}\n{{- messages[idx]['content'] + ' [/INST]' -}}\n{%- endif -%}\n{% elif messages[idx]['role'] == 'system' %}\n{{- '[INST] <<SYS>>\\n' + messages[idx]['content'] + '\\n<</SYS>>\\n\\n' -}}\n{%- elif messages[idx]['role'] == 'assistant' -%}\n{{- ' '  + messages[idx]['content'] + ' ' + eos_token -}}\n{% endif %}\n{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 32768,
+  "pad_token": "<s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}