phi2_multiple-choice

README.md

@@ -35,12 +35,12 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 1
+- learning_rate: 0.0002
+- train_batch_size: 2
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: constant
+- lr_scheduler_type: cosine
 - num_epochs: 10
 
 ### Training results

adapter_config.json

@@ -19,9 +19,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "Wqkv",
     "fc2",
     "out_proj",
+    "Wqkv",
     "fc1"
   ],
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6dff85514638c261f7e7c750a8d406e79cfd9d326c398bb4d66fdb218bfa8b39
+oid sha256:822b3b9e199c3378f82110fded80c734303c562cbc8f49e88fc7277ed117f0ac
 size 83919184

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0463c14bf1391a6b200a6e9959982901127b51d491e60d307093243272eb0ef6
+oid sha256:bc5ae1fee6349437c776dffd73e2c62eb835a290370b283f8319d133a953cbf7
 size 4347