Upload folder using huggingface_hub (#3)

- 85d915d7878324eb59a67628c1209e23c1322d10e19f6ad681cc4a79f3591f7c (12cab331b686ed2c50156a9919b2e41379b86f8d)
- c59e15490e11479cf687c25902a3f517e098af1367d3e2d55a11c4e7f68190e3 (86b73252ccf978342944bec061f8f907e8b72f3b)
- c79dd03db625f10edc3b00811d3888fe5cd41b25598125accc3c1653c35019ee (97dfff44f44445dc3c0d3260c2c07c5897b34b38)
- 3fe323d4120cd3375673453998ab105206ab094cd6f2908ba67a2fe9228057c1 (d9f7a4707252fbfad87ac156fb1025796e0f89ea)
- 32a67da953b8c160569cbcff906becc58cb641bdb51b9ce7deb9916b11ffdb6e (79d7c21aab2eddffc9cc6f318939041850ddf46a)
- 69e958e9f211affa9e70263c567c47e6bb4921bd2e4f9d354a7edcd1f26a21ce (bf75b12df892de7ac6717d12d60c4c8f6408b439)
- 347c3e87f6b06bed1123bae751e04e74131819a3feb6d2a896eae35da53be620 (7078c45d87a8305a5b8f2fc45472c766dcf5af33)
- e475021b2fc4d9e209f0a902d98fbf5461e4c343aba15ff995055c1cfa780fa8 (f300f5a5adca0b40aec6a925d01471da3be29047)
- f53dc08b38ebfd6534ddf627bbad4aee45089cb573f723abda428e83f25657a3 (eed66ae0f101fd79406e7c364f1a56220661547e)
- a80bd4923cb01ddbbb94473baeea41240e0b114ffee98473d143a2d1fcdcaa79 (3f4a0f76a4b4b4c8d3e35276813475449eb6d7e5)
- 87c4b1f661ac695a5431fb281e924520e15e6d72d9ba7477b42a2b7ca7f40043 (4783c1bba0cafc4c8b3c711c83d8e1772135aba0)
- 4bb369a0c183bb678a44bd63a1f3fa5ad27779c84ffa59b8ea203534af067c99 (e22345fa81be134f93de54eba2de31b59fa635bf)
- eeb0be3c0a612b41fd89aee994c5c67311e294a11e16f862a47c48f1c8285c26 (0862f82b608d3b96d75d02ab1e292feff58e339a)
- 088a1b13cbac33bafa445b55cde1c479c6cfff99a47fc736e9850a75c248b043 (f7704f2b7d92a7b02233319d1af1832ffc39724a)

config.json

@@ -23,23 +23,21 @@
   "quantization_config": {
     "_load_in_4bit": true,
     "_load_in_8bit": false,
-    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_compute_dtype": "float32",
     "bnb_4bit_quant_storage": "uint8",
     "bnb_4bit_quant_type": "fp4",
     "bnb_4bit_use_double_quant": false,
     "llm_int8_enable_fp32_cpu_offload": false,
     "llm_int8_has_fp16_weight": false,
-    "llm_int8_skip_modules": [
-      "lm_head"
-    ],
+    "llm_int8_skip_modules": null,
     "llm_int8_threshold": 6.0,
     "load_in_4bit": true,
     "load_in_8bit": false,
     "quant_method": "bitsandbytes"
   },
   "rope_theta": 75000000.0,
-  "torch_dtype": "bfloat16",
-  "transformers_version": "4.39.3",
+  "torch_dtype": "float16",
+  "transformers_version": "4.40.0.dev0",
   "use_cache": true,
   "use_qk_norm": true,
   "vocab_size": 256000

generation_config.json

@@ -3,5 +3,5 @@
   "bos_token_id": 5,
   "eos_token_id": 255001,
   "pad_token_id": 0,
-  "transformers_version": "4.39.3"
+  "transformers_version": "4.40.0.dev0"
 }

model-00001-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24556465c06abb47e07225bf653535e59081099da5319299a294a2cfa03e0af3
+oid sha256:ee321057203258a9356a5fe30deab557b0b7070d1da93153b5dddd66328cf000
 size 6291456144

model-00002-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29e178fbfbfe94571750c29e9b86567aa020cc9d358eeca4c7303aed6c7ddc8a
-size 4841424092
+oid sha256:bc6e0fbf34d0556f2493cfef9af9028f34022d43f93656b809318fef15fc3912
+size 4841585116

model-00003-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73047ab2f7ea8f7c37a879c1d825cb25ed4962809bf32ba5512d8a2858a24820
-size 4990084444
+oid sha256:a9fe7f614fdd5f5c89e073f7ca007a787a6716ef292968b864dda9ec46daf7dc
+size 4990245468

model-00004-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8f2ec69c369998b4ca039735bc2a0d7b8857a0806d40d42fb9c495731ae077c
-size 4975902838
+oid sha256:b40ea27cb60fdaa9929e9a5ca9e6d3778ce2116125303952ea363e45280048a3
+size 4976037032

model-00005-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c856cfa7343ad282452f103bd19798c20a223313d99399f467c128245e46e66
-size 4841448956
+oid sha256:d9fd066feceee09ae3d635a82a5d2e9185c4fbbabe0fc617c7044dd1c6cfc49b
+size 4841609996

model-00006-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91d13810aabccf6566ca9f781103ed8dc302d31faf02469342e66ce1327dbf84
-size 4890967546
+oid sha256:b8f2feab12ad1bb6f33139d770a395cbe54e4487707aa120c3e3308d50fc9c89
+size 4891101741

model-00007-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:933a6f9722b4ce31e4a4449146af7d267f2eb7e9b4756e662ac46886a92d5879
-size 4841448956
+oid sha256:7b13d297600dd703c65d8761ca5f3179f3c46cd1d5635bd08768f4dabbf763fa
+size 4841609996

model-00008-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad477af24261d70a0215b7ec6aa022c04bce81f704abc98dff88cc6781adbe48
-size 4890967546
+oid sha256:2917c4ad99a6ecfbb1c2f2115ca641a89fef4d3c22eced0a3e2fb1e290eb7744
+size 4891101741

model-00009-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca7824fee63f98bdd3f4d474055851792683c70bfeaa5e81f7ea1db0712b9852
-size 4841448956
+oid sha256:b9ea840650e66a38915d6b11411f871d4576b9937abd8d2f1ec1ab2d2cf06a0f
+size 4841609996

model-00010-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2a538ce4cba9a4e5db72968a6bc792ae1ef893f8c727a48d00f7c527121df08
-size 4890967546
+oid sha256:e65ff8d242b12d9604ec233e2d3bd46d5de5cd3e1b9f637a1363fffae57f0026
+size 4891101741

model-00011-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18832ba8fc377cd101182bc1375e1733765b872e7e9ae9545eef4d4516879fe1
-size 4841448956
+oid sha256:097002aa2fc8ce2d7e2e088a81d7c4e203ec739673c12a34c5a234849270d410
+size 4841609996

model-00012-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b3fd6a1c3ed2297cabc3674d8168fbeee07c3e706a5433a8b4cc49967786a01
-size 4890967546
+oid sha256:f4df3d4a2d7e70b27b27eaae3883198af66c75c099f49f5bd372e1486b3e9939
+size 4891101741

model-00013-of-00013.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d56000379065b81ac28d6407242dddae5d7a2478db48fca3e59168def0e7cbf9
-size 2887915698
+oid sha256:352a5e44a0c7257724cf7cbca26f30fe84dbe89b932c67630adef1e07d22bd8d
+size 2887996212

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 62916223616
+    "total_size": 62917927104
   },
   "weight_map": {
     "model.embed_tokens.weight": "model-00001-of-00013.safetensors",
@@ -17,6 +17,7 @@
     "model.layers.0.mlp.up_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.0.mlp.up_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.0.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.0.self_attn.k_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.k_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.k_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -25,6 +26,7 @@
     "model.layers.0.self_attn.o_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.o_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.0.self_attn.q_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.q_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.0.self_attn.q_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -46,6 +48,7 @@
     "model.layers.1.mlp.up_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.1.mlp.up_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.1.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.1.self_attn.k_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.k_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.k_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -54,6 +57,7 @@
     "model.layers.1.self_attn.o_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.o_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.1.self_attn.q_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.q_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.1.self_attn.q_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -75,6 +79,7 @@
     "model.layers.10.mlp.up_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.10.mlp.up_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.10.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.10.self_attn.k_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.k_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.k_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -83,6 +88,7 @@
     "model.layers.10.self_attn.o_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.o_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.10.self_attn.q_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.q_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.10.self_attn.q_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -104,6 +110,7 @@
     "model.layers.11.mlp.up_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.11.mlp.up_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.11.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.11.self_attn.k_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.11.self_attn.k_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.11.self_attn.k_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -112,6 +119,7 @@
     "model.layers.11.self_attn.o_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.11.self_attn.o_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.11.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.11.self_attn.q_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.11.self_attn.q_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.11.self_attn.q_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -133,6 +141,7 @@
     "model.layers.12.mlp.up_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.12.mlp.up_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.12.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.12.self_attn.k_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.k_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.k_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.k_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -141,6 +150,7 @@
     "model.layers.12.self_attn.o_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.o_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.12.self_attn.q_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.q_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.q_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.12.self_attn.q_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -162,6 +172,7 @@
     "model.layers.13.mlp.up_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.13.mlp.up_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.13.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.13.self_attn.k_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.k_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.k_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -170,6 +181,7 @@
     "model.layers.13.self_attn.o_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.o_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.13.self_attn.q_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.q_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.13.self_attn.q_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -191,6 +203,7 @@
     "model.layers.14.mlp.up_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.14.mlp.up_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.14.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.14.self_attn.k_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.k_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.k_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -199,6 +212,7 @@
     "model.layers.14.self_attn.o_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.o_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.14.self_attn.q_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.q_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.14.self_attn.q_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -220,6 +234,7 @@
     "model.layers.15.mlp.up_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.15.mlp.up_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.15.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.15.self_attn.k_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.k_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.k_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -228,6 +243,7 @@
     "model.layers.15.self_attn.o_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.o_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.15.self_attn.q_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.q_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.15.self_attn.q_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -249,6 +265,7 @@
     "model.layers.16.mlp.up_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.16.mlp.up_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.16.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.16.self_attn.k_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.k_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.k_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -257,6 +274,7 @@
     "model.layers.16.self_attn.o_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.o_proj.weight.quant_map": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00004-of-00013.safetensors",
+    "model.layers.16.self_attn.q_norm.weight": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.q_proj.weight.absmax": "model-00004-of-00013.safetensors",
     "model.layers.16.self_attn.q_proj.weight.quant_map": "model-00004-of-00013.safetensors",
@@ -278,6 +296,7 @@
     "model.layers.17.mlp.up_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.17.mlp.up_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.17.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.17.self_attn.k_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.k_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.k_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.k_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -286,6 +305,7 @@
     "model.layers.17.self_attn.o_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.o_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.17.self_attn.q_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.q_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.q_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.17.self_attn.q_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -307,6 +327,7 @@
     "model.layers.18.mlp.up_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.18.mlp.up_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.18.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.18.self_attn.k_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.k_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.k_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -315,6 +336,7 @@
     "model.layers.18.self_attn.o_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.o_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.18.self_attn.q_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.q_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.18.self_attn.q_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -336,6 +358,7 @@
     "model.layers.19.mlp.up_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.19.mlp.up_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.19.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.19.self_attn.k_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.k_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.k_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -344,6 +367,7 @@
     "model.layers.19.self_attn.o_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.o_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.19.self_attn.q_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.q_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.19.self_attn.q_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -365,6 +389,7 @@
     "model.layers.2.mlp.up_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.2.mlp.up_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.2.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.2.self_attn.k_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.k_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.k_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -373,6 +398,7 @@
     "model.layers.2.self_attn.o_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.o_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.2.self_attn.q_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.q_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.2.self_attn.q_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -394,6 +420,7 @@
     "model.layers.20.mlp.up_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.20.mlp.up_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.20.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.20.self_attn.k_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.k_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.k_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -402,6 +429,7 @@
     "model.layers.20.self_attn.o_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.o_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.20.self_attn.q_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.q_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.20.self_attn.q_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -423,6 +451,7 @@
     "model.layers.21.mlp.up_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.21.mlp.up_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.21.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.21.self_attn.k_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.k_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.k_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -431,6 +460,7 @@
     "model.layers.21.self_attn.o_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.o_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.21.self_attn.q_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.q_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.21.self_attn.q_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -452,6 +482,7 @@
     "model.layers.22.mlp.up_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.22.mlp.up_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.22.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.22.self_attn.k_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.k_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.k_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -460,6 +491,7 @@
     "model.layers.22.self_attn.o_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.o_proj.weight.quant_map": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00005-of-00013.safetensors",
+    "model.layers.22.self_attn.q_norm.weight": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.q_proj.weight.absmax": "model-00005-of-00013.safetensors",
     "model.layers.22.self_attn.q_proj.weight.quant_map": "model-00005-of-00013.safetensors",
@@ -481,6 +513,7 @@
     "model.layers.23.mlp.up_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.23.mlp.up_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.23.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.23.self_attn.k_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.k_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.k_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -489,6 +522,7 @@
     "model.layers.23.self_attn.o_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.o_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.23.self_attn.q_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.q_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.23.self_attn.q_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -510,6 +544,7 @@
     "model.layers.24.mlp.up_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.24.mlp.up_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.24.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.24.self_attn.k_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.k_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.k_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -518,6 +553,7 @@
     "model.layers.24.self_attn.o_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.o_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.24.self_attn.q_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.q_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.24.self_attn.q_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -539,6 +575,7 @@
     "model.layers.25.mlp.up_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.25.mlp.up_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.25.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.25.self_attn.k_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.k_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.k_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -547,6 +584,7 @@
     "model.layers.25.self_attn.o_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.o_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.25.self_attn.q_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.q_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.25.self_attn.q_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -568,6 +606,7 @@
     "model.layers.26.mlp.up_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.26.mlp.up_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.26.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.26.self_attn.k_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.k_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.k_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -576,6 +615,7 @@
     "model.layers.26.self_attn.o_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.o_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.26.self_attn.q_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.q_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.26.self_attn.q_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -597,6 +637,7 @@
     "model.layers.27.mlp.up_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.27.mlp.up_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.27.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.27.self_attn.k_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.k_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.k_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -605,6 +646,7 @@
     "model.layers.27.self_attn.o_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.o_proj.weight.quant_map": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00006-of-00013.safetensors",
+    "model.layers.27.self_attn.q_norm.weight": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.q_proj.weight.absmax": "model-00006-of-00013.safetensors",
     "model.layers.27.self_attn.q_proj.weight.quant_map": "model-00006-of-00013.safetensors",
@@ -626,6 +668,7 @@
     "model.layers.28.mlp.up_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.28.mlp.up_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.28.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.28.self_attn.k_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.k_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.k_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.k_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -634,6 +677,7 @@
     "model.layers.28.self_attn.o_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.o_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.28.self_attn.q_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.q_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.q_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.28.self_attn.q_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -655,6 +699,7 @@
     "model.layers.29.mlp.up_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.29.mlp.up_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.29.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.29.self_attn.k_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.k_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.k_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.k_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -663,6 +708,7 @@
     "model.layers.29.self_attn.o_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.o_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.29.self_attn.q_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.q_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.q_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.29.self_attn.q_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -684,6 +730,7 @@
     "model.layers.3.mlp.up_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.3.mlp.up_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.3.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.3.self_attn.k_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.k_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.k_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -692,6 +739,7 @@
     "model.layers.3.self_attn.o_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.o_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.3.self_attn.q_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.q_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.3.self_attn.q_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -713,6 +761,7 @@
     "model.layers.30.mlp.up_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.30.mlp.up_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.30.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.30.self_attn.k_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.k_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.k_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -721,6 +770,7 @@
     "model.layers.30.self_attn.o_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.o_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.30.self_attn.q_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.q_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.30.self_attn.q_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -742,6 +792,7 @@
     "model.layers.31.mlp.up_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.31.mlp.up_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.31.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.31.self_attn.k_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.k_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.k_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.k_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -750,6 +801,7 @@
     "model.layers.31.self_attn.o_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.o_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.31.self_attn.q_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.q_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.q_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.31.self_attn.q_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -771,6 +823,7 @@
     "model.layers.32.mlp.up_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.32.mlp.up_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.32.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.32.self_attn.k_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.k_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.k_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -779,6 +832,7 @@
     "model.layers.32.self_attn.o_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.o_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.32.self_attn.q_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.q_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.32.self_attn.q_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -800,6 +854,7 @@
     "model.layers.33.mlp.up_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.33.mlp.up_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.33.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.33.self_attn.k_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.k_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.k_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.k_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -808,6 +863,7 @@
     "model.layers.33.self_attn.o_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.o_proj.weight.quant_map": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00007-of-00013.safetensors",
+    "model.layers.33.self_attn.q_norm.weight": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.q_proj.weight": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.q_proj.weight.absmax": "model-00007-of-00013.safetensors",
     "model.layers.33.self_attn.q_proj.weight.quant_map": "model-00007-of-00013.safetensors",
@@ -829,6 +885,7 @@
     "model.layers.34.mlp.up_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.34.mlp.up_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.34.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.34.self_attn.k_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.k_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.k_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.k_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -837,6 +894,7 @@
     "model.layers.34.self_attn.o_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.o_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.34.self_attn.q_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.q_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.q_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.34.self_attn.q_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -858,6 +916,7 @@
     "model.layers.35.mlp.up_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.35.mlp.up_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.35.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.35.self_attn.k_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.k_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.k_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.k_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -866,6 +925,7 @@
     "model.layers.35.self_attn.o_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.o_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.35.self_attn.q_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.q_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.q_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.35.self_attn.q_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -887,6 +947,7 @@
     "model.layers.36.mlp.up_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.36.mlp.up_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.36.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.36.self_attn.k_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.k_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.k_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.k_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -895,6 +956,7 @@
     "model.layers.36.self_attn.o_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.o_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.36.self_attn.q_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.q_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.q_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.36.self_attn.q_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -916,6 +978,7 @@
     "model.layers.37.mlp.up_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.37.mlp.up_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.37.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.37.self_attn.k_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.k_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.k_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.k_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -924,6 +987,7 @@
     "model.layers.37.self_attn.o_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.o_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.37.self_attn.q_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.q_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.q_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.37.self_attn.q_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -945,6 +1009,7 @@
     "model.layers.38.mlp.up_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.38.mlp.up_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.38.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.38.self_attn.k_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.k_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.k_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.k_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -953,6 +1018,7 @@
     "model.layers.38.self_attn.o_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.o_proj.weight.quant_map": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00008-of-00013.safetensors",
+    "model.layers.38.self_attn.q_norm.weight": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.q_proj.weight": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.q_proj.weight.absmax": "model-00008-of-00013.safetensors",
     "model.layers.38.self_attn.q_proj.weight.quant_map": "model-00008-of-00013.safetensors",
@@ -974,6 +1040,7 @@
     "model.layers.39.mlp.up_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.39.mlp.up_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.39.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.39.self_attn.k_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.k_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.k_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.k_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -982,6 +1049,7 @@
     "model.layers.39.self_attn.o_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.o_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.39.self_attn.q_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.q_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.q_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.39.self_attn.q_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1003,6 +1071,7 @@
     "model.layers.4.mlp.up_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.4.mlp.up_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.4.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.4.self_attn.k_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.k_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.k_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -1011,6 +1080,7 @@
     "model.layers.4.self_attn.o_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.o_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.4.self_attn.q_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.q_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.4.self_attn.q_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -1032,6 +1102,7 @@
     "model.layers.40.mlp.up_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.40.mlp.up_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.40.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.40.self_attn.k_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.k_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.k_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.k_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1040,6 +1111,7 @@
     "model.layers.40.self_attn.o_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.o_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.40.self_attn.q_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.q_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.q_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.40.self_attn.q_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1061,6 +1133,7 @@
     "model.layers.41.mlp.up_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.41.mlp.up_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.41.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.41.self_attn.k_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.k_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.k_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.k_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1069,6 +1142,7 @@
     "model.layers.41.self_attn.o_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.o_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.41.self_attn.q_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.q_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.q_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.41.self_attn.q_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1090,6 +1164,7 @@
     "model.layers.42.mlp.up_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.42.mlp.up_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.42.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.42.self_attn.k_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.k_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.k_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.k_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1098,6 +1173,7 @@
     "model.layers.42.self_attn.o_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.o_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.42.self_attn.q_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.q_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.q_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.42.self_attn.q_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1119,6 +1195,7 @@
     "model.layers.43.mlp.up_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.43.mlp.up_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.43.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.43.self_attn.k_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.k_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.k_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.k_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1127,6 +1204,7 @@
     "model.layers.43.self_attn.o_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.o_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.43.self_attn.q_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.q_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.q_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.43.self_attn.q_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1148,6 +1226,7 @@
     "model.layers.44.mlp.up_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.44.mlp.up_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.44.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.44.self_attn.k_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.k_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.k_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.k_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1156,6 +1235,7 @@
     "model.layers.44.self_attn.o_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.o_proj.weight.quant_map": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00009-of-00013.safetensors",
+    "model.layers.44.self_attn.q_norm.weight": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.q_proj.weight": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.q_proj.weight.absmax": "model-00009-of-00013.safetensors",
     "model.layers.44.self_attn.q_proj.weight.quant_map": "model-00009-of-00013.safetensors",
@@ -1177,6 +1257,7 @@
     "model.layers.45.mlp.up_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.45.mlp.up_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.45.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.45.self_attn.k_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.k_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.k_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.k_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1185,6 +1266,7 @@
     "model.layers.45.self_attn.o_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.o_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.45.self_attn.q_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.q_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.q_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.45.self_attn.q_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1206,6 +1288,7 @@
     "model.layers.46.mlp.up_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.46.mlp.up_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.46.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.46.self_attn.k_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.k_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.k_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.k_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1214,6 +1297,7 @@
     "model.layers.46.self_attn.o_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.o_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.46.self_attn.q_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.q_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.q_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.46.self_attn.q_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1235,6 +1319,7 @@
     "model.layers.47.mlp.up_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.47.mlp.up_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.47.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.47.self_attn.k_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.k_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.k_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.k_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1243,6 +1328,7 @@
     "model.layers.47.self_attn.o_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.o_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.47.self_attn.q_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.q_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.q_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.47.self_attn.q_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1264,6 +1350,7 @@
     "model.layers.48.mlp.up_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.48.mlp.up_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.48.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.48.self_attn.k_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.k_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.k_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.k_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1272,6 +1359,7 @@
     "model.layers.48.self_attn.o_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.o_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.48.self_attn.q_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.q_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.q_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.48.self_attn.q_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1293,6 +1381,7 @@
     "model.layers.49.mlp.up_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.49.mlp.up_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.49.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.49.self_attn.k_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.k_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.k_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.k_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1301,6 +1390,7 @@
     "model.layers.49.self_attn.o_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.o_proj.weight.quant_map": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00010-of-00013.safetensors",
+    "model.layers.49.self_attn.q_norm.weight": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.q_proj.weight": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.q_proj.weight.absmax": "model-00010-of-00013.safetensors",
     "model.layers.49.self_attn.q_proj.weight.quant_map": "model-00010-of-00013.safetensors",
@@ -1322,6 +1412,7 @@
     "model.layers.5.mlp.up_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.5.mlp.up_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.5.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.5.self_attn.k_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.k_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.k_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -1330,6 +1421,7 @@
     "model.layers.5.self_attn.o_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.o_proj.weight.quant_map": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00002-of-00013.safetensors",
+    "model.layers.5.self_attn.q_norm.weight": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.q_proj.weight.absmax": "model-00002-of-00013.safetensors",
     "model.layers.5.self_attn.q_proj.weight.quant_map": "model-00002-of-00013.safetensors",
@@ -1351,6 +1443,7 @@
     "model.layers.50.mlp.up_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.50.mlp.up_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.50.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.50.self_attn.k_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.k_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.k_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.k_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1359,6 +1452,7 @@
     "model.layers.50.self_attn.o_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.o_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.50.self_attn.q_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.q_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.q_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.50.self_attn.q_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1380,6 +1474,7 @@
     "model.layers.51.mlp.up_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.51.mlp.up_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.51.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.51.self_attn.k_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.k_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.k_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.k_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1388,6 +1483,7 @@
     "model.layers.51.self_attn.o_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.o_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.51.self_attn.q_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.q_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.q_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.51.self_attn.q_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1409,6 +1505,7 @@
     "model.layers.52.mlp.up_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.52.mlp.up_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.52.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.52.self_attn.k_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.k_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.k_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.k_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1417,6 +1514,7 @@
     "model.layers.52.self_attn.o_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.o_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.52.self_attn.q_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.q_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.q_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.52.self_attn.q_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1438,6 +1536,7 @@
     "model.layers.53.mlp.up_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.53.mlp.up_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.53.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.53.self_attn.k_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.k_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.k_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.k_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1446,6 +1545,7 @@
     "model.layers.53.self_attn.o_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.o_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.53.self_attn.q_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.q_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.q_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.53.self_attn.q_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1467,6 +1567,7 @@
     "model.layers.54.mlp.up_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.54.mlp.up_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.54.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.54.self_attn.k_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.k_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.k_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.k_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1475,6 +1576,7 @@
     "model.layers.54.self_attn.o_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.o_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.54.self_attn.q_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.q_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.q_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.54.self_attn.q_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1496,6 +1598,7 @@
     "model.layers.55.mlp.up_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.55.mlp.up_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.55.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.55.self_attn.k_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.k_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.k_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.k_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1504,6 +1607,7 @@
     "model.layers.55.self_attn.o_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.o_proj.weight.quant_map": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00011-of-00013.safetensors",
+    "model.layers.55.self_attn.q_norm.weight": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.q_proj.weight": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.q_proj.weight.absmax": "model-00011-of-00013.safetensors",
     "model.layers.55.self_attn.q_proj.weight.quant_map": "model-00011-of-00013.safetensors",
@@ -1525,6 +1629,7 @@
     "model.layers.56.mlp.up_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.56.mlp.up_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.56.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.56.self_attn.k_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.k_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.k_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.k_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1533,6 +1638,7 @@
     "model.layers.56.self_attn.o_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.o_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.56.self_attn.q_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.q_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.q_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.56.self_attn.q_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1554,6 +1660,7 @@
     "model.layers.57.mlp.up_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.57.mlp.up_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.57.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.57.self_attn.k_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.k_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.k_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.k_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1562,6 +1669,7 @@
     "model.layers.57.self_attn.o_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.o_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.57.self_attn.q_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.q_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.q_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.57.self_attn.q_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1583,6 +1691,7 @@
     "model.layers.58.mlp.up_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.58.mlp.up_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.58.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.58.self_attn.k_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.k_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.k_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.k_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1591,6 +1700,7 @@
     "model.layers.58.self_attn.o_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.o_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.58.self_attn.q_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.q_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.q_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.58.self_attn.q_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1612,6 +1722,7 @@
     "model.layers.59.mlp.up_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.59.mlp.up_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.59.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.59.self_attn.k_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.k_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.k_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.k_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1620,6 +1731,7 @@
     "model.layers.59.self_attn.o_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.o_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.59.self_attn.q_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.q_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.q_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.59.self_attn.q_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1641,6 +1753,7 @@
     "model.layers.6.mlp.up_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.6.mlp.up_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.6.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.6.self_attn.k_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.k_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.k_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.k_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -1649,6 +1762,7 @@
     "model.layers.6.self_attn.o_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.o_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.6.self_attn.q_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.q_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.q_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.6.self_attn.q_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -1670,6 +1784,7 @@
     "model.layers.60.mlp.up_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.60.mlp.up_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.60.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.60.self_attn.k_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.k_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.k_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.k_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1678,6 +1793,7 @@
     "model.layers.60.self_attn.o_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.o_proj.weight.quant_map": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00012-of-00013.safetensors",
+    "model.layers.60.self_attn.q_norm.weight": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.q_proj.weight": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.q_proj.weight.absmax": "model-00012-of-00013.safetensors",
     "model.layers.60.self_attn.q_proj.weight.quant_map": "model-00012-of-00013.safetensors",
@@ -1699,6 +1815,7 @@
     "model.layers.61.mlp.up_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.61.mlp.up_proj.weight.quant_map": "model-00013-of-00013.safetensors",
     "model.layers.61.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00013-of-00013.safetensors",
+    "model.layers.61.self_attn.k_norm.weight": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.k_proj.weight": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.k_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.k_proj.weight.quant_map": "model-00013-of-00013.safetensors",
@@ -1707,6 +1824,7 @@
     "model.layers.61.self_attn.o_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.o_proj.weight.quant_map": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00013-of-00013.safetensors",
+    "model.layers.61.self_attn.q_norm.weight": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.q_proj.weight": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.q_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.61.self_attn.q_proj.weight.quant_map": "model-00013-of-00013.safetensors",
@@ -1728,6 +1846,7 @@
     "model.layers.62.mlp.up_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.62.mlp.up_proj.weight.quant_map": "model-00013-of-00013.safetensors",
     "model.layers.62.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00013-of-00013.safetensors",
+    "model.layers.62.self_attn.k_norm.weight": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.k_proj.weight": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.k_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.k_proj.weight.quant_map": "model-00013-of-00013.safetensors",
@@ -1736,6 +1855,7 @@
     "model.layers.62.self_attn.o_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.o_proj.weight.quant_map": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00013-of-00013.safetensors",
+    "model.layers.62.self_attn.q_norm.weight": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.q_proj.weight": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.q_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.62.self_attn.q_proj.weight.quant_map": "model-00013-of-00013.safetensors",
@@ -1757,6 +1877,7 @@
     "model.layers.63.mlp.up_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.63.mlp.up_proj.weight.quant_map": "model-00013-of-00013.safetensors",
     "model.layers.63.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00013-of-00013.safetensors",
+    "model.layers.63.self_attn.k_norm.weight": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.k_proj.weight": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.k_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.k_proj.weight.quant_map": "model-00013-of-00013.safetensors",
@@ -1765,6 +1886,7 @@
     "model.layers.63.self_attn.o_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.o_proj.weight.quant_map": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00013-of-00013.safetensors",
+    "model.layers.63.self_attn.q_norm.weight": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.q_proj.weight": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.q_proj.weight.absmax": "model-00013-of-00013.safetensors",
     "model.layers.63.self_attn.q_proj.weight.quant_map": "model-00013-of-00013.safetensors",
@@ -1786,6 +1908,7 @@
     "model.layers.7.mlp.up_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.7.mlp.up_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.7.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.7.self_attn.k_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.k_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.k_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.k_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -1794,6 +1917,7 @@
     "model.layers.7.self_attn.o_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.o_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.7.self_attn.q_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.q_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.q_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.7.self_attn.q_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -1815,6 +1939,7 @@
     "model.layers.8.mlp.up_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.8.mlp.up_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.8.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.8.self_attn.k_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.k_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.k_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.k_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -1823,6 +1948,7 @@
     "model.layers.8.self_attn.o_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.o_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.8.self_attn.q_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.q_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.q_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.8.self_attn.q_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -1844,6 +1970,7 @@
     "model.layers.9.mlp.up_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.9.mlp.up_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.9.mlp.up_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.9.self_attn.k_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.k_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.k_proj.weight.quant_map": "model-00003-of-00013.safetensors",
@@ -1852,6 +1979,7 @@
     "model.layers.9.self_attn.o_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.o_proj.weight.quant_map": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.o_proj.weight.quant_state.bitsandbytes__fp4": "model-00003-of-00013.safetensors",
+    "model.layers.9.self_attn.q_norm.weight": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.q_proj.weight.absmax": "model-00003-of-00013.safetensors",
     "model.layers.9.self_attn.q_proj.weight.quant_map": "model-00003-of-00013.safetensors",