add model

added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

config.json

@@ -0,0 +1,82 @@
+{
+  "_name_or_path": "microsoft/Phi-3-mini-4k-instruct",
+  "architectures": [
+    "LlavaPhi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "microsoft/Phi-3-mini-4k-instruct--configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "microsoft/Phi-3-mini-4k-instruct--modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "connector_depth": 3,
+  "connector_only": false,
+  "do_sample": true,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32007,
+  "freeze_mm_mlp_adapter": false,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "image_aspect_ratio": "pad",
+  "image_position": 13,
+  "image_token_len": 576,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "mm_projector_lr": null,
+  "mm_projector_type": "sva",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_sampler_lr": null,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower_aux_list": [
+    "siglip/CLIP-ViT-SO400M-14-384",
+    "openai/clip-vit-large-patch14-336",
+    "facebook/dinov2-giant-res378",
+    "clip-convnext-XXL-multi-stage"
+  ],
+  "mm_vision_tower_aux_token_len_list": [
+    576,
+    576,
+    576,
+    9216
+  ],
+  "mm_vision_tower_lr": null,
+  "model_type": "llava_phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "num_of_vision_sampler_layers": 10,
+  "num_query_group": 1,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "query_num_list": [
+    576
+  ],
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "sliding_window": 2047,
+  "start_of_vision_sampler_layers": 0,
+  "stride_of_vision_sampler_layers": 3,
+  "tie_word_embeddings": false,
+  "tokenizer_model_max_length": 2048,
+  "tokenizer_padding_side": "right",
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "tune_mm_mlp_adapter": false,
+  "unfreeze_mm_vision_tower": false,
+  "use_cache": true,
+  "use_mm_proj": true,
+  "vision_hidden_size": 1024,
+  "vision_tower_aux_token_len_list": [
+    576,
+    576,
+    576,
+    9216
+  ],
+  "vocab_size": 32064
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": [
+    32000,
+    32001,
+    32007
+  ],
+  "pad_token_id": 32000,
+  "transformers_version": "4.39.3"
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35b459d96fc289d8d6b6116cbecd8365cd20548de5a73e8bf333bb9fe7da7daa
+size 4961868984

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f075be59fe4c49e32317ed5bcddf0660b96a93c19dde35556894c9a60a7f39a
+size 4983111176

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bea235bb9ee60f063c51ca4fadc79b511241f8b5e340fe85cec0fce5375805af
+size 4995731336

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe06640cec81de87f73a460649611f08b47220742bd2dad5e68ba500e0137ec9
+size 1437285416

model.safetensors.index.json

@@ -0,0 +1,687 @@
+{
+  "metadata": {
+    "total_size": 16377909248
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.image_newline": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.qkv_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.qkv_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.self_attn.qkv_proj.weight": "model-00001-of-00004.safetensors",
+    "model.mm_projector.0.bias": "model-00003-of-00004.safetensors",
+    "model.mm_projector.0.weight": "model-00003-of-00004.safetensors",
+    "model.mm_projector.2.bias": "model-00003-of-00004.safetensors",
+    "model.mm_projector.2.weight": "model-00003-of-00004.safetensors",
+    "model.mm_projector_aux_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_0.2.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_0.2.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_0.3.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_0.3.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_1.2.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_1.2.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_1.3.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_1.3.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_2.2.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_2.2.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_2.3.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_2.3.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_3.2.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_3.2.weight": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_3.3.bias": "model-00004-of-00004.safetensors",
+    "model.mm_projector_aux_3.3.weight": "model-00004-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors",
+    "model.vision_query": "model-00001-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.1.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_0.layers.2.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.0.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.1.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.2.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.3.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.4.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.5.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.6.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.7.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.8.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.k_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.q_proj.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.q_proj.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.q_proj.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_0.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_0.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_0.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_1.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_1.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_1.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_2.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_2.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_2.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_3.0.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_3.0.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.cross_attn.v_proj_3.1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.norm.bias": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.norm.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.pos_embed_3": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.proj_context.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.proj_in.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.proj_out.linear_1.weight": "model-00004-of-00004.safetensors",
+    "model.vision_sampler_layers.9.layers.0.proj_out.linear_2.weight": "model-00004-of-00004.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<unk>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') %}{{'<|user|>' + '\n' + message['content'] + '<|end|>' + '\n' + '<|assistant|>' + '\n'}}{% elif (message['role'] == 'assistant') %}{{message['content'] + '<|end|>' + '\n'}}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "legacy": false,
+  "model_max_length": 2048,
+  "pad_token": "<unk>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e77b47cc02120974f45fc0e5a5698d66184f6bba9d1775a023d4134f30d6c961
+size 5944