KhaiQuang/ApecGPT-v2

README.md

@@ -40,7 +40,7 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
-- training_steps: 1500
+- training_steps: 2000
 
 ### Training results
 

adapter_config.json

@@ -20,9 +20,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "attn.out_proj",
-    "attn.Wqkv",
     "ffn.up_proj",
+    "attn.Wqkv",
+    "attn.out_proj",
     "ffn.down_proj"
   ],
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b526b65a06f50a39d31b138021ce26d08f2e223ef6f21f0a627452e9dbaece11
+oid sha256:7d10fc64414806a4fd247ca982936b759af88d3722ec757805fd1ac1bd9d8d8b
 size 100698208

runs/Apr23_04-48-00_77f10519ad9f/events.out.tfevents.1713847681.77f10519ad9f.582.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e716722448f4fb192e4a06e0fa6b421cae3efb1ea245aa0c6f54ad268ddec5c
+size 48277

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de8dff27144e19028cbd8590fd2cb485cf98d184634a07c7fcadb4dbc786ae91
+oid sha256:3ca3aa1893e9ce504047e0397a480bff0253b5bc629190d557db4c013bf83167
 size 4984