Training in progress, step 100

adapter_config.json

@@ -10,23 +10,23 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
+  "lora_alpha": 32,
   "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
-    "k_proj",
-    "q_proj",
     "o_proj",
-    "gate_proj",
+    "down_proj",
     "v_proj",
-    "up_proj"
+    "q_proj",
+    "up_proj",
+    "k_proj",
+    "gate_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:a7cc1037b4043e1ffed555e5b5a3c2965a0d8d8048a8cb2c2c39f06907e95e5f
-size 19644912
+oid sha256:052df175bf4eee183ec1fb92bb878b0cbf612f43f65e96af51e1b332fef7c848
+size 39256960

runs/Jun10_12-29-30_user-HP-Z8-Fury-G5-Workstation-Desktop-PC/events.out.tfevents.1717990183.user-HP-Z8-Fury-G5-Workstation-Desktop-PC.3621.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2486a6d55c399a67b9a51d32a73e9ef1ce53884aaf9410a4607c44863a98d896
+size 9976

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd5877e43d5753535e758abb02b6480d721ce9ea601957b5deb1ae2918c39674
-size 5240
+oid sha256:3edc5c56647434dd64e8429c57952e35e9a5b3bd9fbe0871db5df341e3c2fc32
+size 5368