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: 16
 - eval_batch_size: 16
 - seed: 42
 - distributed_type: multi-GPU
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 40
+- total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - num_epochs: 1

adapter_config.json

@@ -16,9 +16,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
     "q_proj",
     "k_proj",
+    "o_proj",
     "v_proj"
   ],
   "task_type": "CAUSAL_LM"

runs/Feb04_00-09-28_r2d2-devbox-fqvnh/events.out.tfevents.1738627784.r2d2-devbox-fqvnh.17579.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa23907217066abe27755afee5a0534e3c7d7deb4e93547cae5719d620ccac93
-size 4703
+oid sha256:b16cc8c4754ccd8b2343f0c62c0aeafed19e169e4c6922491092d70ace8b825c
+size 4857

runs/Feb04_00-10-45_r2d2-devbox-fqvnh/events.out.tfevents.1738627861.r2d2-devbox-fqvnh.18026.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c5fdf77c97c328631b8ecfd1d20a277b6fb65b239566e48aa2c64f3ff0351a9
+size 4703

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0e54077e4bdcdd3996c6f99df5cf25d02b3a6726850bc66f5302f35ed8bcc44
+oid sha256:a3972cee2946370190289439aab4182858c0c8b33c9729de0d54ccc9dd049719
 size 5816