Training in progress, step 25

adapter_config.json

@@ -16,17 +16,17 @@
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 32,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "q_proj",
-    "down_proj",
     "k_proj",
+    "q_proj",
     "o_proj",
     "up_proj",
-    "gate_proj"
+    "gate_proj",
+    "v_proj",
+    "down_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:b3509b50bcd5244ceaea99a8668143d5b7b3dbf87f9036bf66e77fadd2b869c0
-size 25271744
+oid sha256:380942d61ada848609deb225e4e6f543bfb4e266deadde334d47e7cd742106ec
+size 100966336

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:904f7c41a3e205f7ef7cb6db9ec1f789efbcc60a14457fc0f78a11b3e3eddee9
-size 6776
+oid sha256:ad180fe191fac96daa079ae5cf3b983175865cf949ded264830df0cc29889a2c
+size 6712