Training in progress, epoch 0

adapter_config.json

@@ -16,12 +16,20 @@
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "gate_proj",
+    "up_proj",
+    "visual_proj",
     "q_proj",
-    "v_proj"
+    "down_proj",
+    "embed_tokens",
+    "v_proj",
+    "o_proj",
+    "wte",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94ffa54e8bf17a08a1b1bd44d753583beb20d281c7e769040a2817589b73b089
-size 4372840
+oid sha256:35bafcfd10cc4c5541bb4e44d7273729746664a105347fe7e21c1dc15c921407
+size 545570000

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcf0c1f9407f9c686993aea6d3928c4af97ab8540359c9991ea9ec6cc2b93938
+oid sha256:acff61c2b37d65c324b821e6318585624abfe412e4d74c05093c6f5d3caf57e4
 size 5560