Upload 8 files

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "./checkpoints/llavaqwen-2b-other",
+  "_name_or_path": "./checkpoints/llava-qwen-2b-otherdata-V",
   "architectures": [
     "MoELLaVAQWenForCausalLM"
   ],
@@ -8,9 +8,9 @@
     "AutoConfig": "Qwen/Qwen-1_8B--configuration_qwen.QWenConfig",
     "AutoModelForCausalLM": "Qwen/Qwen-1_8B--modeling_qwen.QWenLMHeadModel"
   },
-  "bf16": true,
+  "bf16": false,
   "emb_dropout_prob": 0.0,
-  "fp16": false,
+  "fp16": true,
   "fp32": false,
   "freeze_mm_mlp_adapter": false,
   "hidden_size": 2048,
@@ -79,7 +79,6 @@
   "num_attention_heads": 16,
   "num_hidden_layers": 24,
   "onnx_safe": null,
-  "pad_token_id": 151646,
   "rotary_emb_base": 10000,
   "rotary_pct": 1.0,
   "scale_attn_weights": true,
@@ -88,14 +87,14 @@
   "tie_word_embeddings": false,
   "tokenizer_class": "QWenTokenizer",
   "tokenizer_padding_side": "right",
-  "torch_dtype": "bfloat16",
-  "transformers_version": "4.36.2",
+  "torch_dtype": "float16",
+  "transformers_version": "4.31.0",
   "tune_mm_mlp_adapter": false,
   "use_cache": true,
   "use_cache_kernel": false,
   "use_cache_quantization": false,
   "use_dynamic_ntk": true,
-  "use_flash_attn": true,
+  "use_flash_attn": false,
   "use_logn_attn": true,
   "use_mm_proj": true,
   "video_global_proj": false,

generation_config.json

@@ -11,5 +11,5 @@
   ],
   "top_k": 0,
   "top_p": 0.8,
-  "transformers_version": "4.36.2"
+  "transformers_version": "4.31.0"
 }

pytorch_model-00001-of-00002.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:381272c21a15dcdd5f3c682ce1802ebff5d3782554bcc60cc3214af1448d68b2
+size 9959655631

pytorch_model-00002-of-00002.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5ec2dd5e1a8ee4604aa5ac0cb9b5a217166bce7ca23f830fecb97ba9a17589e
+size 3497114837

pytorch_model.bin.index.json

@@ -0,0 +1,717 @@
+{
+  "metadata": {
+    "total_size": 13456506880
+  },
+  "weight_map": {
+    "lm_head.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.0.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.0.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.1.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.10.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.11.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.12.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.13.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.14.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.15.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.16.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.17.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.18.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.19.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.2.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.20.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.21.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.21.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.mlp.w1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.21.mlp.w2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.22.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.c_attn.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.c_attn.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.attn.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.ln_1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.ln_2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.mlp.c_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.mlp.w1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.23.mlp.w2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.h.3.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.3.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.4.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.5.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.6.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.7.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.0.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.0.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.0.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.1.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.1.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.1.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.2.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.2.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.2.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.3.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.3.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.experts.deepspeed_experts.3.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.8.mlp.deepspeed_moe.gate.wg.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.c_attn.bias": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.c_attn.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.attn.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.ln_1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.ln_2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.mlp.c_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.mlp.w1.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.h.9.mlp.w2.weight": "pytorch_model-00001-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.embeddings.class_embedding": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.embeddings.patch_embedding.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.embeddings.position_embedding.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.0.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.1.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.10.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.11.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.17.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.18.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.19.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.2.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.20.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.21.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.22.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.23.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.3.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.4.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.5.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.6.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.7.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.8.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.layer_norm1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.layer_norm1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.layer_norm2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.layer_norm2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.encoder.layers.9.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.post_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.post_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.pre_layrnorm.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.image_tower.image_tower.vision_model.pre_layrnorm.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.ln_f.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.mm_projector.image_spatial_proj.0.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.mm_projector.image_spatial_proj.0.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.mm_projector.image_spatial_proj.2.bias": "pytorch_model-00002-of-00002.bin",
+    "transformer.mm_projector.image_spatial_proj.2.weight": "pytorch_model-00002-of-00002.bin",
+    "transformer.wte.weight": "pytorch_model-00001-of-00002.bin"
+  }
+}

special_tokens_map.json

@@ -1,17 +1,6 @@
 {
-  "eos_token": {
-    "content": "<|endoftext|>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
+  "bos_token": "<|extra_1|>",
+  "eos_token": "<|endoftext|>",
   "pad_token": "<|extra_0|>",
-  "unk_token": {
-    "content": "<|extra_0|>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  }
+  "unk_token": "<|extra_0|>"
 }

tokenizer_config.json

@@ -1,5 +1,4 @@
 {
-  "added_tokens_decoder": {},
   "auto_map": {
     "AutoTokenizer": [
       "Qwen/Qwen-1_8B--tokenization_qwen.QWenTokenizer",
@@ -7,11 +6,8 @@
     ]
   },
   "clean_up_tokenization_spaces": true,
-  "eos_token": "<|endoftext|>",
   "model_max_length": 2048,
-  "pad_token": "<|extra_0|>",
   "padding_side": "right",
   "tokenizer_class": "QWenTokenizer",
-  "unk_token": "<|extra_0|>",
   "use_fast": false
 }