Training in progress, epoch 1

adapter_config.json

@@ -7,14 +7,15 @@
   "base_model_name_or_path": "google/vit-base-patch16-224-in21k",
   "bias": "none",
   "fan_in_fan_out": false,
+  "gradient_accumulation_steps": 4,
   "inference_mode": true,
   "init_lora_weights": "svd",
   "layer_replication": null,
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
-  "lora_dropout": 0.15,
+  "lora_alpha": 1,
+  "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": [
@@ -27,11 +28,13 @@
   "revision": null,
   "scheduler": "cos",
   "target_modules": [
+    "dense",
+    "key",
     "value",
     "query"
   ],
   "task_type": null,
-  "total_steps": 175,
+  "total_steps": 43,
   "use_dora": false,
   "use_rslora": false
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e358a6fe38535467edcdfaa5dd5094487234aa0c3d4e11d796b9d5d15cb1c15
-size 4028948
+oid sha256:83c2cfdcddd5279cda0e6fc05c8ee33465daf5c9c5d5be4b47153f55645bd6c2
+size 16992908

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a05cde11187b38db5295436050eab62b1229d79c1651782ca9960a4acb607f15
+oid sha256:a10b251852b85906f087e8a98e34874d6b03e54bb6e55fc8a14edf9c4eda17fd
 size 5304