Training in progress, step 7000

adapter_config.json

@@ -13,24 +13,24 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 32,
+  "lora_alpha": 16,
   "lora_bias": false,
   "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 16,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v",
     "wo",
-    "k",
-    "q",
-    "wi_1",
     "wi_0",
-    "o"
+    "o",
+    "v",
+    "wi_1",
+    "q",
+    "k"
   ],
   "task_type": "SEQ_2_SEQ_LM",
   "trainable_token_indices": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29888d6bb2be75dd6e2a81fe9169e6e5f7d56c2a7132aa83dd296bba2d28cb3b
-size 13628856
+oid sha256:cde04cd2a61be20c7106ba279c33a1c478f6e9e4fa59212d2e2afa40e1ddc305
+size 6845168

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f15b19231f2ba3e94843a225bbda5d88491e69856348aceb2cb19286ffd7b52
-size 7032
+oid sha256:22c1eac10003e02e542944770c1f8ca0540fc11dd9061b6fa4af5997d7a9e983
+size 7160