Training in progress, step 30

adapter_config.json

@@ -12,24 +12,21 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 192,
+  "lora_alpha": 128,
   "lora_bias": false,
-  "lora_dropout": 0.04,
+  "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 96,
+  "r": 64,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "down_proj",
-    "up_proj",
-    "gate_proj",
     "q_proj",
-    "o_proj",
-    "k_proj"
+    "v_proj",
+    "k_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a5188898ce40bb194c5629240084f580d980f51ef9516f9c8a04bb1e4886d12
-size 1200148368
+oid sha256:50910ce9aff057f6d46b561800d5bb27ae18b300d6d583d4146761e95806adc2
+size 205551296

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f45f3d5a01fe5ec3b5d4bcda2e954510c518137417ea404b85746971950e6ce
+oid sha256:439c1536f7b1d5c596c3640a33dcc4cd16eafa8a6436ae49d4491382d1146b9a
 size 5560