Model save

README.md

@@ -34,11 +34,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 6e-05
-- train_batch_size: 4
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 16
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 64
 - optimizer: Use adamw_hf with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 3

adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "q_proj",
     "up_proj",
-    "down_proj",
+    "q_proj",
     "gate_proj",
-    "o_proj",
-    "v_proj"
+    "v_proj",
+    "down_proj",
+    "k_proj",
+    "o_proj"
   ],
   "task_type": null,
   "use_dora": true,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1722f6dac1e1ed8540b99df2276c1da436921cdcec06d0ad26468427713b1cd6
+oid sha256:fe1d0fb76b2e6856c4622847685edf013db90020624041931b3de9f043374e79
 size 51802928

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e961a0009640256e26a3c3a818d3b931622674907f303d90030377e62d78b2f4
+oid sha256:d3537a01668977a56c41e9d156bd3cc5869bf96cd826c11b6768d9c0ecbb232c
 size 5240