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: 16
+- train_batch_size: 13
 - eval_batch_size: 16
 - seed: 42
 - distributed_type: multi-GPU
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 64
+- total_train_batch_size: 52
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - num_epochs: 1

adapter_config.json

@@ -16,10 +16,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "k_proj",
     "o_proj",
-    "v_proj"
+    "v_proj",
+    "q_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

runs/Feb04_00-12-10_r2d2-devbox-fqvnh/events.out.tfevents.1738627946.r2d2-devbox-fqvnh.18473.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b1c6ba54d7b1ad3fb47af5dca83525ca6d998ef825e78ed7f0fd12ec30b81f4
+size 4703

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3972cee2946370190289439aab4182858c0c8b33c9729de0d54ccc9dd049719
+oid sha256:13bbc3463190dd66ef2467e45eed5bd6ee52a065a550b44c520b532aa53223de
 size 5816