Training in progress, step 150

adapter_config.json

@@ -10,22 +10,22 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 128,
-  "lora_dropout": 0.2,
+  "lora_alpha": 16,
+  "lora_dropout": 0.0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 64,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "lm_head",
-    "word_embeddings",
     "query_key_value",
-    "dense_h_to_4h",
+    "word_embeddings",
     "dense",
-    "dense_4h_to_h"
+    "dense_h_to_4h",
+    "dense_4h_to_h",
+    "lm_head"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9416216736ab96969ca153b7227c1057aba72a2eab0cd5342017cd1aa8fb5bdb
-size 2284875672
+oid sha256:3a3e20ddb40ef1c2b4b80be403a68896d0c8960b6a3d8944099ccf1bbf1b4a6f
+size 2083942096

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:707c86cf4fdd0646439a88306d68c5f5f607150b508e479a7e53f5fa4d44df2f
-size 5496
+oid sha256:981bea758e0e0d96856dae23fee6f6b8ca842aab6dc422b6d95c8ab288849b99
+size 5368