Training in progress, step 300

README.md

@@ -1,17 +1,17 @@
 ---
-base_model: Qwen/Qwen2.5-VL-32B-Instruct
+base_model: Qwen/Qwen2.5-VL-7B-Instruct
 library_name: transformers
 model_name: anti_aesthetics_captioning
 tags:
 - generated_from_trainer
-- sft
 - trl
+- sft
 licence: license
 ---
 
 # Model Card for anti_aesthetics_captioning
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-VL-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-32B-Instruct).
+This model is a fine-tuned version of [Qwen/Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/3dsmile/huggingface/runs/epb6vhi4) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/3dsmile/huggingface/runs/4n52rfyh) 
 
 
 This model was trained with SFT.

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "Qwen/Qwen2.5-VL-32B-Instruct",
+  "base_model_name_or_path": "Qwen/Qwen2.5-VL-7B-Instruct",
   "bias": "none",
   "corda_config": null,
   "eva_config": null,
@@ -13,7 +13,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 32,
+  "lora_alpha": 64,
   "lora_bias": false,
   "lora_dropout": 0.0,
   "megatron_config": null,
@@ -21,14 +21,15 @@
   "modules_to_save": null,
   "peft_type": "LORA",
   "qalora_group_size": 16,
-  "r": 32,
+  "r": 64,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "v_proj",
+    "k_proj",
     "o_proj",
-    "k_proj"
+    "qkv",
+    "v_proj",
+    "q_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0d5da870d48a8f905b015d6e2efcd452a686163e847df272b3d891f39157a97
-size 268512816
+oid sha256:8d4c646c251e57b1c0db2d3dd76a524ed1dece3cc8d2031eedd7abcf5375836c
+size 203466072

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db86ecd266c79166b251c06cf57c50214857fa4fe82b2bedb895efeb20ee9014
+oid sha256:ed701b2b37e1bca68597c2d604aee0f225b006e0ed7e46fcbd34bee44b9a8e60
 size 5777