Model save

README.md

@@ -42,8 +42,10 @@ The following hyperparameters were used during training:
 - eval_batch_size: 8
 - seed: 4
 - distributed_type: multi-GPU
+- num_devices: 2
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 16
+- total_train_batch_size: 32
+- total_eval_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - num_epochs: 1

adapter_config.json

@@ -19,12 +19,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "q_proj",
-    "up_proj",
+    "v_proj",
     "o_proj",
+    "q_proj",
+    "k_proj",
     "down_proj",
-    "v_proj"
+    "up_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

runs/May16_15-44-36_c842148f8542/events.out.tfevents.1715874324.c842148f8542.21570.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f88c67cfb4b6c56c505d7e4f4820e2b0c7029d990cc8f229adb5503ee0433ac
+size 4477

step_0/adapter_config.json

@@ -19,12 +19,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "q_proj",
-    "up_proj",
+    "v_proj",
     "o_proj",
+    "q_proj",
+    "k_proj",
     "down_proj",
-    "v_proj"
+    "up_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

step_0/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0972caf6dfbe417b15985b1fa88b70e41a7682ab205c10ce25d2feef0352bdc5
+oid sha256:037cb93649ce2c00bfe42f23882c655e67e006c8997cb25bf1e587e809060da7
 size 5688

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0972caf6dfbe417b15985b1fa88b70e41a7682ab205c10ce25d2feef0352bdc5
+oid sha256:037cb93649ce2c00bfe42f23882c655e67e006c8997cb25bf1e587e809060da7
 size 5688