Upload 9 files

README.md

@@ -0,0 +1,9 @@
+---
+library_name: peft
+---
+## Training procedure
+
+### Framework versions
+
+
+- PEFT 0.4.0

Vit-lora/README.md

@@ -0,0 +1,6 @@
+## Training procedure
+
+### Framework versions
+
+
+- PEFT 0.4.0

Vit-lora/adapter_config.json

@@ -0,0 +1,28 @@
+{
+  "auto_mapping": {
+    "base_model_class": "CLIPVisionTower",
+    "parent_library": "llava.model.multimodal_encoder.clip_encoder"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 64,
+  "lora_dropout": 0.2,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "revision": null,
+  "target_modules": [
+    "fc2",
+    "fc1",
+    "v_proj",
+    "k_proj",
+    "out_proj",
+    "q_proj"
+  ],
+  "task_type": null
+}

Vit-lora/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2fee99628d040795f47dea12c58b261562b0659ee84a4b0733f833dcb1b529c
+size 28418445

adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./checkpoints/ShareGPT4V-7B_Pretrained_vit-large336-l12_vicuna-7b-v1.5",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 256,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj",
+    "down_proj",
+    "up_proj",
+    "k_proj",
+    "o_proj",
+    "gate_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e398ad10211ce2fd96e83c534e1e6aa25a9fe97976156b28de07d414f0ef69f
+size 668211021

config.json

@@ -0,0 +1,43 @@
+{
+  "_name_or_path": "./checkpoints/ShareGPT4V-7B_Pretrained_vit-large336-l12_vicuna-7b-v1.5",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "freeze_mm_mlp_adapter": false,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "image_aspect_ratio": "pad",
+  "image_grid_pinpoints": null,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "loss_alpha": 0.5,
+  "max_position_embeddings": 4096,
+  "mm_hidden_size": 1024,
+  "mm_projector_lr": 2e-05,
+  "mm_projector_type": "cluster",
+  "mm_use_im_patch_token": false,
+  "mm_use_im_start_end": false,
+  "mm_vision_select_feature": "patch",
+  "mm_vision_select_layer": -2,
+  "mm_vision_tower": "Lin-Chen/ShareGPT4V-7B_Pretrained_vit-large336-l12",
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "num_latents": 256,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "qav_loss": false,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.31.0",
+  "tune_mm_mlp_adapter": false,
+  "use_cache": true,
+  "use_mm_proj": true,
+  "vision_tower_lr": 0.0002,
+  "vocab_size": 32000
+}

non_lora_trainables.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:224f33d1aa6c047c4aea18aa70dbc0b41d9b2e1db7a28fb7c99db80015df20e0
+size 48265979