Upload model weights (#1)

- 8336e7ed285f289e7a44c8c620b32bdfd9cf381aa022515bfec07fd3bb507d2e (0696d762b979850890027579ed2f419bf4e83b51)
- fadae7300d605177f911a6bcbd16f2cdd7c41c1802bf39ffe4735758cd1ace00 (c84f7453037444dba3cf17cfd9a4cab3b74b25ba)
- b33c44af268f4e058f992018e40e99674a1ee63307550dd39eb6c925626189e1 (223044653ed24ab6c1047ca0706d4db4f6f7749a)
- 2b3841b784bd55a38c8265e2bf4a63bd4659cd47e502cd26c916d9fe20bcbacc (e1c27cdfe78dbbd99fe2b0d9e71598e04095ac7b)
- e78858cf26a19416e6c4e1ec185bdd6e7d9b0e6caa69ce8d0ebbc53e0f5278d9 (be29be3c0c80ca6d1aa07d14c50f2b49efa01f8c)
- 20d0522dcd123d74b5b60c62542644f5b155d159d741547cf6c359e7a4a79f3f (0acf83459db2f666d6bcb49d6899433e2ffd2c44)
- ef7f23fc3774aac936d6abf66ffb083536124016ecef188b7e264f1dd2b075e4 (6a84d3189583e2331f01e4c23467f7ceb9144d4c)
- c26c12a59422a4bc24cd3a864f755d871dcb595ca002e1bf1ca4b88c0739221c (b42dcbc66418260b1b46632f01c03842deaa684f)
- 501f13533d13701193425644f3b26e60e85cd526df7651772655ba655d866dd1 (a84e3e28a6df875256ddaa8ef7431a2398f2b571)
- d6bcc276874edf91ad9ea76aa3abde2569d370ecd6b7739bab72b3f58d675bfc (583dcfaa54008aafb19f5857192e5d3b3b57fdd5)
- d7a831482ba836c1511d3d121cc291ba651ac18ed318cf99a4599959769df1d9 (8f69ea8b6ebf245b0daae0e8350a97c9ef71418c)
- d2c9c4ddb98190c695566a486b669e7f84434938fb954b74aa85193268d5db82 (6cae612c440b4d324548f35561979579096f54a8)
- d97496a8a7eb064c505c11edb501506feaac62ed5913ad99e8a637862ea895c1 (aa8c74518e0ab155ca0933579d0f029583d34637)
- ed2096e531d7e0ecfa964f7207a7c526ea5021d24da5cdaf9d4f207c631ed9b3 (5e8170650d80bf37dcb66ff431e7297cb5e4e0f3)
- f8802046184dbce1ae8f2b24bb601a8fc27989a1bf9410be10dc53b169548242 (3e85c274660928d4c2113a4083ddf49f1cd0fbe3)
- 379f5c800ef8ffd6ff45dc3021f6b68aacb3bdbe9bc6407875bb006d6def133f (14f038a09340c8aa20ad7c1b9fb782a018754307)
- 5a7038a3d3988fc998fae735d8cb55352481515528a88e6aa4ff3ba85af4246b (f74df2ee27459843d4d901f0bdb06ee7552b37b3)
- d645e0f98d7af9cef1ff51e2094e89208561f70f454fa7745cf35cd8f1e22398 (79aef9d4e068b80ef97d1c300ea80a4846cb42bb)
- 8f9b1c589040729200e0f096318bf60ef9645f8a02aeecebb9f9101acc274d6b (4fed5cb4a4cb2922d4f54704310c98a74405cd1d)
- 2152b1b2f2140480fef2d8d84f858e2f12a5fc28650f182a3042f5a62754707f (d8a778c2df668c748dfb538cb1cb35354a64c9ce)
- e74163e3c2aabbc60ee4c8225f553f6328a3e79f88a7f42dca3c66b0b8f95bd3 (cea475c7a35bc642746c6e7bf47daec7c85bea93)
- 85ef59ed97cdef5b575a3f2766985003a36eae494efe7bc2e4bc946017044862 (995fed992e6dd7a43627d5d89be97538965e3eec)

config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "/home/llm/mergequant/models/stellarbright-xwin-v3",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 4096,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 120,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.35.2",
+  "use_cache": false,
+  "vocab_size": 32000
+}

model-00001-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d75077ceab0b5be6e55e312f29fa9a39c8355b8d8a339412d303fc6561a56c2
+size 9852622800

model-00002-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad20db8b057970a42632ce182d533f23a97253ed9cbc8dc1ee5001056ea3a56b
+size 9965852192

model-00003-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8146632203ae50114cfb8cde9630c5f4a62af44dd2acf2942120e292306846eb
+size 9798096912

model-00004-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb80187c917b2c5d08e8b30bee972e863c6f4bd7a28247e7d0827e6e1b82784c
+size 9663846080

model-00005-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779f4f0be89e5369a83f01949963245533e6e4cd1988493cdb7e64b275b5967b
+size 9630340896

model-00006-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa063178b3960e36866e9ae117a9d9f828157b1d70c8ff659b644b25876d52fe
+size 9798047424

model-00007-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c618e5820d8503843518bb74b09f4d49f9a74f9ea84f8d3375960dcce2fabc7
+size 9982679048

model-00008-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfa39005bcb01a1ef82001f690a7ec39807c51f5bfb78b70b27915b77d2cebae
+size 9613514784

model-00009-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5a4cd7ec63b9213745ef9b2dcadefdef5afc40d692ea003606754791e9a3805
+size 9680655696

model-00010-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e20732fab5f69a1e89e5209a8ef07e0979887b8153c4658f5f0938e38b930376
+size 9932298264

model-00011-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49aeb59b955676571d3b41955bdd1d3e647772c6ba437becf5adb0fed1d4e7ff
+size 9647085232

model-00012-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d935cb8b251e824731580a8275f93a2636e6db1087ead76ec3cfbb33a0870acc
+size 9798063920

model-00013-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d262bd924a9d402d6c5b9eae4818344af4176979bca45b10b2bfcc55f0d04dcd
+size 9798096904

model-00014-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7267995e8de67f85fc34b51a20eab0e121452efadafd45b3814186451d2241ac
+size 9798113400

model-00015-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f94c827f192a390acf1c533eb53fa80e6cf14c1da6f74043d1d1c6d0e27a6ee5
+size 9798080424

model-00016-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b788264fe672f02d268eecaafc26386d804679236d4945acf0e984c0b358d7a
+size 9680655720

model-00017-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fa99b8bb274c7ccecce15394f459936a7a728e962989ffc40c7b52a9a467299
+size 9932298280

model-00018-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:840078651594217c2b60bf779a40c3053dd4e82ef9360e504dec5ad692baea47
+size 9647101760

model-00019-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:548497171b6f01166919406cec168623a5a0747aeec9c9ec916be5ee2309a5cc
+size 9798047440

model-00020-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bade24ddf1cf8ac517ca1b91f55a01cfb77def4971dd91c13cc1f92d364239e
+size 9982679072

model-00021-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dea23d815a1c01f6c03623aec57ca3be86bc19006b821b3456d54fefe1ac8623
+size 9835812568

model-00022-of-00022.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce7aeb4bd57e8f159df240d71cb1d25a5f929bba82d87ce82ae19f130665d933
+size 771785520

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.3"}, "weight_map": {"model.embed_tokens.weight": "model-00001-of-00022.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.5.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00001-of-00022.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00022.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00022.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00022.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00022.safetensors", "model.layers.10.input_layernorm.weight": "model-00001-of-00022.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.11.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.6.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.9.input_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00002-of-00022.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00022.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00002-of-00022.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00022.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00022.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00022.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00022.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00022.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00003-of-00022.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00003-of-00022.safetensors", "model.layers.17.input_layernorm.weight": "model-00003-of-00022.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.18.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.input_layernorm.weight": "model-00004-of-00022.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00004-of-00022.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.23.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.24.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.25.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00005-of-00022.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00005-of-00022.safetensors", "model.layers.28.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00005-of-00022.safetensors", "model.layers.40.input_layernorm.weight": "model-00005-of-00022.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.41.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.42.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.29.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00006-of-00022.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00006-of-00022.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00006-of-00022.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00006-of-00022.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00006-of-00022.safetensors", "model.layers.30.input_layernorm.weight": "model-00006-of-00022.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00006-of-00022.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00006-of-00022.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.31.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.34.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.43.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.44.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00007-of-00022.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00007-of-00022.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00007-of-00022.safetensors", "model.layers.45.input_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00007-of-00022.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00007-of-00022.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00007-of-00022.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00007-of-00022.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00007-of-00022.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.46.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.47.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.35.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.36.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00008-of-00022.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00008-of-00022.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00008-of-00022.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00008-of-00022.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00008-of-00022.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00008-of-00022.safetensors", "model.layers.37.input_layernorm.weight": "model-00008-of-00022.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00008-of-00022.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.48.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.49.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.50.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00009-of-00022.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00009-of-00022.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00009-of-00022.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00009-of-00022.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00009-of-00022.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00009-of-00022.safetensors", "model.layers.51.input_layernorm.weight": "model-00009-of-00022.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00009-of-00022.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.52.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.80.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.81.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.82.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00010-of-00022.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00010-of-00022.safetensors", "model.layers.83.input_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00010-of-00022.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00010-of-00022.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00010-of-00022.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00010-of-00022.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00010-of-00022.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00010-of-00022.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.85.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.56.input_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00011-of-00022.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00011-of-00022.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00011-of-00022.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00011-of-00022.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00011-of-00022.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00011-of-00022.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00011-of-00022.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00011-of-00022.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.86.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.87.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.88.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.89.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00012-of-00022.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00012-of-00022.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00012-of-00022.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00012-of-00022.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00012-of-00022.safetensors", "model.layers.90.input_layernorm.weight": "model-00012-of-00022.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00012-of-00022.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00012-of-00022.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.91.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.57.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.58.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.input_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00013-of-00022.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00013-of-00022.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00013-of-00022.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00013-of-00022.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00013-of-00022.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00013-of-00022.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.92.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.93.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.94.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.95.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.96.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00014-of-00022.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00014-of-00022.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00014-of-00022.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00014-of-00022.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00014-of-00022.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00014-of-00022.safetensors", "model.layers.97.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.62.input_layernorm.weight": "model-00014-of-00022.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00014-of-00022.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.63.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.64.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00015-of-00022.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00015-of-00022.safetensors", "model.layers.98.input_layernorm.weight": "model-00015-of-00022.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00015-of-00022.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.99.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.100.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.101.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00016-of-00022.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00016-of-00022.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00016-of-00022.safetensors", "model.layers.102.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00016-of-00022.safetensors", "model.layers.67.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00016-of-00022.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00016-of-00022.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.68.input_layernorm.weight": "model-00016-of-00022.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00016-of-00022.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.69.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.70.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.103.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00017-of-00022.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00017-of-00022.safetensors", "model.layers.104.input_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00017-of-00022.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00017-of-00022.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00017-of-00022.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00017-of-00022.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00017-of-00022.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00017-of-00022.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.105.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.106.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.107.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.108.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.72.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.73.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00018-of-00022.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00018-of-00022.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00018-of-00022.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00018-of-00022.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00018-of-00022.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00018-of-00022.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00018-of-00022.safetensors", "model.layers.74.input_layernorm.weight": "model-00018-of-00022.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.75.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.109.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.110.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00019-of-00022.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00019-of-00022.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00019-of-00022.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00019-of-00022.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00019-of-00022.safetensors", "model.layers.111.input_layernorm.weight": "model-00019-of-00022.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00019-of-00022.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00019-of-00022.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.114.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.76.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.77.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.78.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00020-of-00022.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00020-of-00022.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00020-of-00022.safetensors", "model.layers.79.input_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00020-of-00022.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00020-of-00022.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00020-of-00022.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00020-of-00022.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00020-of-00022.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "lm_head.weight": "model-00021-of-00022.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00021-of-00022.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.input_layernorm.weight": "model-00021-of-00022.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00021-of-00022.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00022-of-00022.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00022-of-00022.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00022-of-00022.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00022-of-00022.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00022-of-00022.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00022-of-00022.safetensors", "model.norm.weight": "model-00022-of-00022.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "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": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "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>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "max_length": 2048,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "stride": 0,
+  "tokenizer_class": "LlamaTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}