Training in progress, step 150

adapter_config.json

@@ -10,22 +10,21 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 32,
-  "lora_dropout": 0.1,
+  "lora_alpha": 16,
+  "lora_dropout": 0.2,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": [
     "lm_head"
   ],
   "peft_type": "LORA",
-  "r": 16,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
+    "q_proj",
     "k_proj",
-    "o_proj",
-    "q_proj"
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc3c2d065d0e83c1128ff07b7b1356bf0bd380afc6bb886a60d6bb434fff5283
-size 1130395064
+oid sha256:7e81660a5e55019447836a6013d3ea0aa5c04c4d9f1d33afe72df3e4a7997421
+size 1103779584

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c167a5959afcf2631a587e2c79af9dc4334a99f07862c928036bfa576ad08c6b
+oid sha256:db0b1a8501da2a8e506e8ae3417c049c4e15fc8737da5bf526cb315ba1d54160
 size 5496