Training in progress, step 500

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "cyberagent/calm2-7b",
+  "base_model_name_or_path": "cyberagent/calm2-7b-chat-dpo-experimental",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -20,14 +20,14 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "k_proj",
-    "down_proj",
     "v_proj",
+    "lm_head",
     "up_proj",
+    "q_proj",
+    "k_proj",
     "o_proj",
     "gate_proj",
-    "lm_head"
+    "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:a5efb1e1ddf7e0ff38fe678f573f5512df8ceb3db13e38bf8c9b3a380f10ed8b
+oid sha256:43ccc3a8e8cd74a31931fbd09d98cccfad24d1aa2002a6f9cfff961633098153
 size 1147580432

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:afd6574bdbbcd3ccae3dcb1b0dc558d57e6ecdaa1121d8129de69e55b6c846b4
-size 4027
+oid sha256:0880f6b7bbc476c7173def8993d9af557db8eca15baa0a562a7ee618e4816c0f
+size 4539