your-username/mistral_finetuned

README.md

@@ -1,4 +1,5 @@
 ---
+license: apache-2.0
 library_name: peft
 tags:
 - trl
@@ -17,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4235
+- Loss: 0.3764
 
 ## Model description
 
@@ -36,20 +37,21 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0004
+- learning_rate: 5e-05
 - train_batch_size: 4
 - eval_batch_size: 4
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 1
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.4375        | 1.0   | 1915 | 0.4235          |
+| 0.4035        | 1.0   | 1915 | 0.3892          |
+| 0.3797        | 2.0   | 3830 | 0.3764          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,10 +20,14 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "q_proj",
+    "o_proj",
+    "up_proj",
     "k_proj",
+    "lm_head",
+    "down_proj",
     "v_proj",
-    "q_proj",
-    "o_proj"
+    "gate_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:be6d7c77af8b1d9db600cad4f2ccf373ea22bbca935c0cdbcb08f18ab87d25f4
-size 54560368
+oid sha256:215e942ac19cefaf6701ceab7ba55d24d996aaee9ea33d573e9b0bfcfdb70f54
+size 694431312

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac417a93b84651927a04faf83c654f4b3ca06e11bf68257c043dc230ba0aa2e6
+oid sha256:b778b3ecd8a3450648bd496eb386d5f4fc0b3c5738e9c589eae6a7595f643565
 size 4984