Model save

README.md

@@ -34,12 +34,12 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 10
+- train_batch_size: 11
 - eval_batch_size: 16
 - seed: 42
 - distributed_type: multi-GPU
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 40
+- total_train_batch_size: 44
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - num_epochs: 1

adapter_config.json

@@ -17,9 +17,9 @@
   "revision": null,
   "target_modules": [
     "k_proj",
-    "v_proj",
     "o_proj",
-    "q_proj"
+    "q_proj",
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

runs/Feb04_00-25-46_r2d2-devbox-fqvnh/events.out.tfevents.1738628762.r2d2-devbox-fqvnh.21647.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:06e204bca73e32a41ee03a845c5615964d1eeb3de60eb7bae5dad2cc0f42f631
-size 4703
+oid sha256:79d1aabe4ba057c9ad48b16dd23ceda436566eeb45456e5f41772c19b70aa239
+size 4857

runs/Feb04_00-26-48_r2d2-devbox-fqvnh/events.out.tfevents.1738628824.r2d2-devbox-fqvnh.22093.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fbfbafba53ad7676d497f6aa6ac5a261ca1b657b56ea075e19070bfe012db07
+size 4703

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d953a3b3de3f4b60c1c39fb860e4f69bbd9340413413b46f658bc13ff229d5f8
+oid sha256:8cb086fa0f352a7cdaf6af7173b7ed288f2f20adf12b97a9cf104524dceb24a8
 size 5816