fix(pad zero) pad intermediate_size to 29696 to make sure quantized model can use 8 tensor-parallel in vllm

config.json

@@ -1,20 +1,16 @@
 {
-  "_name_or_path": "Qwen/Qwen2-VL-72B-Instruct-GPTQ-Int8",
+  "_name_or_path": "Qwen/Qwen2-VL-72B-Instruct-315-fix-dim-pad",
   "architectures": [
     "Qwen2VLForConditionalGeneration"
   ],
   "attention_dropout": 0.0,
   "bos_token_id": 151643,
   "eos_token_id": 151645,
-  "vision_start_token_id": 151652,
-  "vision_end_token_id": 151653,
-  "vision_token_id": 151654,
-  "image_token_id": 151655,
-  "video_token_id": 151656,
   "hidden_act": "silu",
   "hidden_size": 8192,
+  "image_token_id": 151655,
   "initializer_range": 0.02,
-  "intermediate_size": 29568,
+  "intermediate_size": 29696,
   "max_position_embeddings": 32768,
   "max_window_layers": 80,
   "model_type": "qwen2_vl",
@@ -48,17 +44,15 @@
   "transformers_version": "4.45.0.dev0",
   "use_cache": true,
   "use_sliding_window": false,
+  "video_token_id": 151656,
   "vision_config": {
-    "depth": 32,
-    "embed_dim": 1280,
-    "mlp_ratio": 4,
-    "num_heads": 16,
-    "in_chans": 3,
     "hidden_size": 8192,
-    "patch_size": 14,
-    "spatial_merge_size": 2,
-    "spatial_patch_size": 14,
-    "temporal_patch_size": 2
+    "in_chans": 3,
+    "model_type": "qwen2_vl",
+    "spatial_patch_size": 14
   },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
   "vocab_size": 152064
 }

model-00002-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08bb28260e24d622e27668e6a03d8af143d0f5623151f23d452aa41073e1bc0d
-size 3919769080
+oid sha256:22b21f6727ea22cc69edecb38bfb2939b7f5501b617a36eda68355f2e0b2ee50
+size 3933722616

model-00003-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:beab19aede9e0ae43e22f33ddea2a687f10e88a7ab3111d1438125514f7d35b5
-size 3842331544
+oid sha256:493b6a9b2df51300de7c77c59db0a4a1c0e1d56e562229f763a333be82a9500f
+size 3856284568

model-00004-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:591982fc0dda4f43769267a265d7e435e650fcfb479c2041fca4ef6fc7ac4bea
-size 3997052216
+oid sha256:bc1b2bcc5ccc001827901e93d244641b7fc466107f8a41afbc7efa2cb97e33f2
+size 3933653192

model-00005-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1b44a2b18f45435952333819aefdefe11245302757280ace2fe9aa8b389635b
-size 3842417184
+oid sha256:110c935df7e411c8f74f2967aaca9e6a4de82025e75257d23c5c322c47ba4593
+size 3933722760

model-00006-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d6488e4420bc01faef1cccfc3e756dbfe1680a152a12caa7c0fae3e45bd8b93
-size 3842331680
+oid sha256:51dcd557a8c049e4a71651c8ed636dd28aa04fc7cd9ae1bb0089894fb9fe2e7a
+size 3856284704

model-00007-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec85660e46ecbaba6ed872889e810a7aa0bc089d9c4e4353fdc425b7fec9c18e
-size 3997052264
+oid sha256:63a2ebb51123b1726a9bd68b0773cd1c3dc31cf8be36551b74624767ef716968
+size 3933653240

model-00008-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4b5dbda7c3fe2e807e04af67e3ee65a9c1e0cd37b5b1087841d06589e4da919
-size 3842417184
+oid sha256:ffcc710199307f40b8e0854233c57561da443fe33300b00fcc401a3d7a94420c
+size 3933722760

model-00009-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:236b33745d06773bccf078a38ec563dbc4037ab7ec755c0fa6e53e2aa4a78d6a
-size 3842331680
+oid sha256:d8dca5803cb00b40867bef50951ab8c15f9c10cd7487f164ae5dffbbd3893914
+size 3856284704

model-00010-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fff317816b3185918f59740dcdc5df80b4f7c5f353972810d5e84ce383443b0
-size 3997052264
+oid sha256:081e6a9f411358b7a91b3444f4ae49534d884232e05e6d9b420d6339e3d25662
+size 3933653240

model-00011-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d09b7131a4c6a25b375b04bccdaf35957a473d89ffe6d83677e7b91c75ed0b78
-size 3842417184
+oid sha256:4f1893bc6b77673e185c8bcfcb10aacc9f1c63785522cafb6c090250919ff1ee
+size 3933722760

model-00012-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fea53411a91d9e46b0a9ff2554037aef559a67f659af7cb098c7ce33c5ed5fe7
-size 3842331680
+oid sha256:7ebcfd57658001265ac2e935bf6c323f2d850fe61d6589abe80a5a327733befe
+size 3856284704

model-00013-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d19f003f3c9f479d4c2b080af61ff23249c7272702ffa84c6242a02cf98ab924
-size 3997052264
+oid sha256:81bd0841831608340f38cb807e72c5bc624f8ac2e991c12dbc84b74c1bb2e4cd
+size 3933653240

model-00014-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:823bd58a92d46ba9fa089713df5221ee29f426d54a81ad9067d95a96313ba149
-size 3842417184
+oid sha256:3a19deab70a4fba0534f53ae393720b47e8a22ae9c3b87509ec4c936438b86de
+size 3933722760

model-00015-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31197c33385fe298b2f0b173ca0e6480a2e416831ccd9a5360a2e766f5460142
-size 3842331680
+oid sha256:02cf34da9b4ed556bde70032e3afe4c3d8caa3a7145c9ea8fef2977624383ee6
+size 3856284704

model-00016-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d436c46d4337b997699d408321d9d82d91d6db340b558e7c341711a04f8b0af4
-size 3997052264
+oid sha256:07ba4cad0397cdeef6c2c7d3c37e7923913c778a9efdfeecb84bc8f25c6ea625
+size 3933653240

model-00017-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b12be6e85fee44720a86d20c856787b7310b8e3af59ff76d22fc2af9a880df3
-size 3842417184
+oid sha256:a3257b539618933c2dbaa57d0b133815bd167415fedf06f0ce76f7aebc3b9bd9
+size 3933722760

model-00018-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11369c3cb0a914e500ac4a329d585da6f6bbe9645cdd95c1253269b37e0f8637
-size 3842331680
+oid sha256:d32113a8a028c7ee42a5695cd0ca766ef8615f1b7732e4e28f14dd22f061887c
+size 3856284704

model-00019-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d4c2a4025a94114b8edfddeab86bac8c1910da695a585a6474d8cda0ad6fd66
-size 3997052264
+oid sha256:f663fddc7364084668196c5fb56c1a635822257e8340b79eed4eb1dd6c4932be
+size 3933653240

model-00020-of-00021.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c19bda47e855ee77e8262790cfb532ab285a7b98cbe0a22cf6248630bd69dc98
-size 1642529048
+oid sha256:680110343579090179e68f0fc5133fe79cd7589525dee02f6a63dbd52957a700
+size 1726321024

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 78269271040
+    "total_size": 78526868480
   },
   "weight_map": {
     "lm_head.weight": "model-00021-of-00021.safetensors",
@@ -193,16 +193,16 @@
     "model.layers.13.self_attn.o_proj.qweight": "model-00004-of-00021.safetensors",
     "model.layers.13.self_attn.o_proj.qzeros": "model-00004-of-00021.safetensors",
     "model.layers.13.self_attn.o_proj.scales": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.q_proj.bias": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.q_proj.g_idx": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.q_proj.qweight": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.q_proj.qzeros": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.q_proj.scales": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.v_proj.bias": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.v_proj.g_idx": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.v_proj.qweight": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.v_proj.qzeros": "model-00004-of-00021.safetensors",
-    "model.layers.13.self_attn.v_proj.scales": "model-00004-of-00021.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.q_proj.g_idx": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.q_proj.qweight": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.q_proj.qzeros": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.q_proj.scales": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.v_proj.g_idx": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.v_proj.qweight": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.v_proj.qzeros": "model-00005-of-00021.safetensors",
+    "model.layers.13.self_attn.v_proj.scales": "model-00005-of-00021.safetensors",
     "model.layers.14.input_layernorm.weight": "model-00005-of-00021.safetensors",
     "model.layers.14.mlp.down_proj.g_idx": "model-00005-of-00021.safetensors",
     "model.layers.14.mlp.down_proj.qweight": "model-00005-of-00021.safetensors",
@@ -655,16 +655,16 @@
     "model.layers.26.self_attn.o_proj.qweight": "model-00007-of-00021.safetensors",
     "model.layers.26.self_attn.o_proj.qzeros": "model-00007-of-00021.safetensors",
     "model.layers.26.self_attn.o_proj.scales": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.q_proj.bias": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.q_proj.g_idx": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.q_proj.qweight": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.q_proj.qzeros": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.q_proj.scales": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.v_proj.bias": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.v_proj.g_idx": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.v_proj.qweight": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.v_proj.qzeros": "model-00007-of-00021.safetensors",
-    "model.layers.26.self_attn.v_proj.scales": "model-00007-of-00021.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.q_proj.g_idx": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.q_proj.qweight": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.q_proj.qzeros": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.q_proj.scales": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.v_proj.g_idx": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.v_proj.qweight": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.v_proj.qzeros": "model-00008-of-00021.safetensors",
+    "model.layers.26.self_attn.v_proj.scales": "model-00008-of-00021.safetensors",
     "model.layers.27.input_layernorm.weight": "model-00008-of-00021.safetensors",
     "model.layers.27.mlp.down_proj.g_idx": "model-00008-of-00021.safetensors",
     "model.layers.27.mlp.down_proj.qweight": "model-00008-of-00021.safetensors",
@@ -1117,16 +1117,16 @@
     "model.layers.39.self_attn.o_proj.qweight": "model-00010-of-00021.safetensors",
     "model.layers.39.self_attn.o_proj.qzeros": "model-00010-of-00021.safetensors",
     "model.layers.39.self_attn.o_proj.scales": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.q_proj.bias": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.q_proj.g_idx": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.q_proj.qweight": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.q_proj.qzeros": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.q_proj.scales": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.v_proj.bias": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.v_proj.g_idx": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.v_proj.qweight": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.v_proj.qzeros": "model-00010-of-00021.safetensors",
-    "model.layers.39.self_attn.v_proj.scales": "model-00010-of-00021.safetensors",
+    "model.layers.39.self_attn.q_proj.bias": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.q_proj.g_idx": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.q_proj.qweight": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.q_proj.qzeros": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.q_proj.scales": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.v_proj.bias": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.v_proj.g_idx": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.v_proj.qweight": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.v_proj.qzeros": "model-00011-of-00021.safetensors",
+    "model.layers.39.self_attn.v_proj.scales": "model-00011-of-00021.safetensors",
     "model.layers.4.input_layernorm.weight": "model-00003-of-00021.safetensors",
     "model.layers.4.mlp.down_proj.g_idx": "model-00002-of-00021.safetensors",
     "model.layers.4.mlp.down_proj.qweight": "model-00002-of-00021.safetensors",
@@ -1612,16 +1612,16 @@
     "model.layers.52.self_attn.o_proj.qweight": "model-00013-of-00021.safetensors",
     "model.layers.52.self_attn.o_proj.qzeros": "model-00013-of-00021.safetensors",
     "model.layers.52.self_attn.o_proj.scales": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.q_proj.bias": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.q_proj.g_idx": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.q_proj.qweight": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.q_proj.qzeros": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.q_proj.scales": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.v_proj.bias": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.v_proj.g_idx": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.v_proj.qweight": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.v_proj.qzeros": "model-00013-of-00021.safetensors",
-    "model.layers.52.self_attn.v_proj.scales": "model-00013-of-00021.safetensors",
+    "model.layers.52.self_attn.q_proj.bias": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.q_proj.g_idx": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.q_proj.qweight": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.q_proj.qzeros": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.q_proj.scales": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.v_proj.bias": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.v_proj.g_idx": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.v_proj.qweight": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.v_proj.qzeros": "model-00014-of-00021.safetensors",
+    "model.layers.52.self_attn.v_proj.scales": "model-00014-of-00021.safetensors",
     "model.layers.53.input_layernorm.weight": "model-00014-of-00021.safetensors",
     "model.layers.53.mlp.down_proj.g_idx": "model-00014-of-00021.safetensors",
     "model.layers.53.mlp.down_proj.qweight": "model-00014-of-00021.safetensors",
@@ -2074,16 +2074,16 @@
     "model.layers.65.self_attn.o_proj.qweight": "model-00016-of-00021.safetensors",
     "model.layers.65.self_attn.o_proj.qzeros": "model-00016-of-00021.safetensors",
     "model.layers.65.self_attn.o_proj.scales": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.q_proj.bias": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.q_proj.g_idx": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.q_proj.qweight": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.q_proj.qzeros": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.q_proj.scales": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.v_proj.bias": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.v_proj.g_idx": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.v_proj.qweight": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.v_proj.qzeros": "model-00016-of-00021.safetensors",
-    "model.layers.65.self_attn.v_proj.scales": "model-00016-of-00021.safetensors",
+    "model.layers.65.self_attn.q_proj.bias": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.q_proj.g_idx": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.q_proj.qweight": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.q_proj.qzeros": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.q_proj.scales": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.v_proj.bias": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.v_proj.g_idx": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.v_proj.qweight": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.v_proj.qzeros": "model-00017-of-00021.safetensors",
+    "model.layers.65.self_attn.v_proj.scales": "model-00017-of-00021.safetensors",
     "model.layers.66.input_layernorm.weight": "model-00017-of-00021.safetensors",
     "model.layers.66.mlp.down_proj.g_idx": "model-00017-of-00021.safetensors",
     "model.layers.66.mlp.down_proj.qweight": "model-00017-of-00021.safetensors",
@@ -2536,16 +2536,16 @@
     "model.layers.78.self_attn.o_proj.qweight": "model-00019-of-00021.safetensors",
     "model.layers.78.self_attn.o_proj.qzeros": "model-00019-of-00021.safetensors",
     "model.layers.78.self_attn.o_proj.scales": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.q_proj.bias": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.q_proj.g_idx": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.q_proj.qweight": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.q_proj.qzeros": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.q_proj.scales": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.v_proj.bias": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.v_proj.g_idx": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.v_proj.qweight": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.v_proj.qzeros": "model-00019-of-00021.safetensors",
-    "model.layers.78.self_attn.v_proj.scales": "model-00019-of-00021.safetensors",
+    "model.layers.78.self_attn.q_proj.bias": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.q_proj.g_idx": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.q_proj.qweight": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.q_proj.qzeros": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.q_proj.scales": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.v_proj.bias": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.v_proj.g_idx": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.v_proj.qweight": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.v_proj.qzeros": "model-00020-of-00021.safetensors",
+    "model.layers.78.self_attn.v_proj.scales": "model-00020-of-00021.safetensors",
     "model.layers.79.input_layernorm.weight": "model-00020-of-00021.safetensors",
     "model.layers.79.mlp.down_proj.g_idx": "model-00020-of-00021.safetensors",
     "model.layers.79.mlp.down_proj.qweight": "model-00020-of-00021.safetensors",