lora-midm-7b-nsmc-understanding

README.md

@@ -41,7 +41,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.03
-- training_steps: 1000
+- training_steps: 1500
 - mixed_precision_training: Native AMP
 
 ### Training results
@@ -54,11 +54,3 @@ The following hyperparameters were used during training:
 - Pytorch 2.1.0+cu118
 - Datasets 2.15.0
 - Tokenizers 0.15.0
-
-### accuracy
-Accuracy: 0.902
-
-Confusion Matrix:
-          Predicted 0  Predicted 1
-Actual 0          434           58
-Actual 1           40          468

adapter_config.json

@@ -19,9 +19,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "c_attn",
     "c_proj",
-    "c_fc"
+    "c_fc",
+    "c_attn"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0eb12ce63d73cf4f2fe773a73c6052b92be17aef53c68bbdb4c7471e8e4a7bba
+oid sha256:fd9d4bfc47c7faec5e9566428fd3db21d668370b9d9ab7fe77582ba65f30cf8f
 size 67010784

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3dd25a991cf6c7d5956de6bbd14c0423c1da7c6d1ffe423a8625e334575972de
+oid sha256:8094d30e1f07593bf8c89ec511b02896b6c90cf9eb81e5f5519ca0d4bc690d33
 size 4664