Model save

README.md

@@ -18,15 +18,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6907
-- Rewards/chosen: 0.0077
-- Rewards/rejected: 0.0025
-- Rewards/accuracies: 0.0472
-- Rewards/margins: 0.0052
-- Logps/rejected: -11.0078
-- Logps/chosen: -7.5850
-- Logits/rejected: -2.2114
-- Logits/chosen: -2.2106
+- Loss: 0.6931
+- Rewards/chosen: 0.0
+- Rewards/rejected: 0.0
+- Rewards/accuracies: 0.0
+- Rewards/margins: 0.0
+- Logps/rejected: -8.1759
+- Logps/chosen: -5.6437
+- Logits/rejected: -2.2170
+- Logits/chosen: -2.2168
 
 ## Model description
 
@@ -54,7 +54,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- training_steps: 5
+- training_steps: 1
 - mixed_precision_training: Native AMP
 
 ### Training results

adapter_config.json

@@ -19,10 +19,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
     "q_proj",
-    "v_proj",
-    "k_proj"
+    "o_proj",
+    "k_proj",
+    "v_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01eff22f270c17d6a16cc359decf3316f37d553aa0632d6211fcd6c7a38470a6
+oid sha256:a3452cef1e3289d786650c8637a888bfeabe74d15bf9a42d89e2cd7de7c0f993
 size 27297032

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a69cb15297afce38a7ecff1f96dc6e9d8a134c2fddf7a65c5382256f3bfe8c25
+oid sha256:5e3ef142ff969fa6655db0c09c77ad43010cacec98d66a168e060189531be034
 size 5112