Training in progress, step 2

README.md

@@ -4,9 +4,9 @@ library_name: transformers
 model_name: trl-demo-qwen
 tags:
 - generated_from_trainer
-- trl
 - sft
 - hf_jobs
+- trl
 licence: license
 ---
 

adapter_config.json

@@ -15,7 +15,7 @@
   "loftq_config": {},
   "lora_alpha": 16,
   "lora_bias": false,
-  "lora_dropout": 0.05,
+  "lora_dropout": 0.0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
@@ -25,8 +25,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "v_proj"
+    "v_proj",
+    "q_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdf31a2bf9c15f763a03dbd6ffbd44a10f51eb6c287184a14e2c03717a05eb12
+oid sha256:ec6a8e56096bf9ea31070ee159c99ae60ed2aff184f731bfab5b2e69e0b41fc9
 size 2175168

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:465f0a24bf6573689ee36e250903fcb5cf352389ce5958d89704b30e00d08a97
-size 6289
+oid sha256:8d7701e4a3d9cc161021bd426ca85dee9ba4ab5f3cb488430f710ac2e2a4eef1
+size 6225