Upload folder using huggingface_hub

config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "TomGrc/FusionNet_34Bx2_MoE_v0.1",
+  "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": 200000,
+  "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": 0,
+  "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": "4.37.2",
+  "use_cache": true,
+  "vocab_size": 64000
+}

model-00001-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44c0e1adaf4c9a1f7516c7912e8e1ab74ad01661828d123a0802717eafd837af
+size 3891099408

model-00002-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cd8583022d4b96e1b90a5fa879eb9d4b500db8d6e33842575bccfc8b82fec29
+size 3816818368

model-00003-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eaea011538f7cb7570d3c1bb2d1fed4a592463948fc6f76438cd6cb8d4c19563
+size 3816818384

model-00004-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67a6113c5f865a39367f8b3f624dcf70d8c61f01ca6bb0ec91d0876950c02fbe
+size 3816818384

model-00005-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08c5c6f8f6fad06523432abe9ae4a969094716c592fca803cca09dda95df732a
+size 3816818384

model-00006-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6247757c6c47507c818be1d332e4bfb485d02d6f72407c5e018e798ac9f3d7bd
+size 3816818384

model-00007-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b45682939c7cb79aabbb305e95c766f2d2b60887d699af3f0333e8e29995b3fe
+size 3816818384

model-00008-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33e8daa2e4bf32350475066f7183ed7097fbc8a983d290c12bc4959e3f2feb8e
+size 3816818384

model-00009-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c50acf70b22a2d47c81ea5d1a91658af12d887679ffa53e114eb065561946b74
+size 3816818384

model-00010-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e550badb0d50b5ca915cefeb2b3988607d2218cc0c78adb6e97b24d02405b063
+size 3816818384

model-00011-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf10f972ed4aaa9df01192a85d4f1b838f20aaca4575d3652df78f487abcd5f3
+size 3816818368

model-00012-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22d7e1c832b2e68e6f7087c4a6fddf5d6b619d67f3da64a5b5f60457a4b2232c
+size 3816818384

model-00013-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5a4ab3d0f92a98daeca7d1d33535760e80e22fc7174ab034891438dfcbf5daa
+size 3816818384

model-00014-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68161962114807d4e7d0a06f43caa89dafb90b14628aa0648778991c09c4dba3
+size 3816818384

model-00015-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b69761f55695f41c9f03732abf0facd82eeebd0c761e79eed466322705ebc1d
+size 3816818384

model-00016-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ebac1dee422f8f80f675ff3ead214dba8e393cddf26e3847a05073478d91681
+size 3816818384

model-00017-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e28ecddce8f1c9413c044506b87abe124529987e8302f121f4b41c132f490aa
+size 3816818384

model-00018-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c02b87ecc88ef22afbf6f4ff5505c3204f29970ec11c6a365b2ee6faa346894
+size 3816818384

model-00019-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e59580d22b4ba774a7e2feb954f1df20d35b503de2401d1e4dea895b5f5bba2c
+size 3816818384

model-00020-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c37e1ae205ee2fe11c86f157de19e8e688696a2a28fdc886c97d8cf4f68b819d
+size 3816818368

model-00021-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbf7b036fe0d332623792b5012735c55683ba368f589dfcae9ffa0df93aab856
+size 3816818376

model-00022-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd48892da33c5d14852fa0f8dae9e49e600dd037059c9cf4292a53ac67785b2a
+size 3816818384

model-00023-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1867cc01bcf4059517facc206208c506d670ef7efd09552f14f03097ebb745d3
+size 3816818384

model-00024-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71273cbc8586e19c366d4eb5fd1f8ed6c5568c1fb1d42facbeb4649567f26d85
+size 3816818384

model-00025-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1b039db05b4f67c9e4e1b8f6eb1782353ac88fcde8f3fa3c61fd55c84dfcb27
+size 3816818384

model-00026-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e79ccd6d101002c058254c2de6dc0b5c02d63758eefd4f50f708343b34ff8d7
+size 3816818384

model-00027-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3870d18bd10ab3436f5d828565fa5e6de31302886665a665ace3fd23c37003d1
+size 3816818384

model-00028-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d159d9bca735207dc261c28ace5d313447fe39b4251b6cc4b3c1dfa902d75b5
+size 3816818384

model-00029-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41bc4783410fbd165edbde3e7857cb1c4b10249ca22ba1913fdf819eca8239ef
+size 3979168888

model-00030-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd70a03e6686f05ead0eff5d43a481d58eb31b5afc59b3729f8d73ad5658010f
+size 3992990288

model-00031-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98d065753c7034db5911bdf7d38ef680f82b3d2b4a4fa4a41ae0be0c802fe21c
+size 3934265416

model-00032-of-00032.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9225da2c8038b47ff84c8aae1a015a1d01b7c9542a4b997cdd349b20fc62f163
+size 2776262560

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4"}, "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 +1,43 @@
+{
+  "add_bos_token": false,
+  "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": 1024,
+  "pad_token": "<s>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}