End of training

README.md

@@ -18,13 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6333
-- Eval/rewards/chosen: -0.0590
-- Eval/logps/chosen: -193.4149
-- Eval/rewards/rejected: -0.0831
-- Eval/logps/rejected: -230.6999
-- Eval/rewards/margins: 0.0241
-- Eval/kl: 0.0070
+- Loss: 0.7265
+- Eval/rewards/chosen: 3.6123
+- Eval/logps/chosen: -167.4125
+- Eval/rewards/rejected: 3.7149
+- Eval/logps/rejected: -178.0628
+- Eval/rewards/margins: -0.1026
+- Eval/kl: 33.2160
 
 ## Model description
 
@@ -43,10 +43,10 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
+- learning_rate: 0.0001
 - train_batch_size: 1
 - eval_batch_size: 2
-- seed: 1234
+- seed: 5678
 - gradient_accumulation_steps: 8
 - total_train_batch_size: 8
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
@@ -56,12 +56,12 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss |        |
-|:-------------:|:-----:|:----:|:---------------:|:------:|
-| 0.9245        | 0.96  | 12   | 0.6312          | 0.0820 |
-| 0.7535        | 2.0   | 25   | 0.6338          | 0.0186 |
-| 0.8857        | 2.96  | 37   | 0.6339          | 0.0081 |
-| 0.6603        | 3.84  | 48   | 0.6333          | 0.0070 |
+| Training Loss | Epoch | Step | Validation Loss |         |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|
+| 0.5619        | 0.96  | 12   | 0.6911          | 17.2746 |
+| 0.5748        | 2.0   | 25   | 0.7189          | 25.3931 |
+| 0.3537        | 2.96  | 37   | 0.7288          | 32.8920 |
+| 0.0845        | 3.84  | 48   | 0.7265          | 33.2160 |
 
 
 ### Framework versions

adapter_config.json

@@ -20,10 +20,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "v_proj",
+    "o_proj",
     "q_proj",
-    "o_proj"
+    "v_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52c9c3364fd1316381e7991b31c09240c05e51dea35e9542595351bc811a744f
+oid sha256:336e0a459c2b4b68cbde3083903a3fe637029eb9b98999c03f264e1ed7582613
 size 27297544

metrics.jsonl

@@ -6,3 +6,7 @@
 {"epoch": 2.0, "precision": 0.4857142843265306, "recall": 0.9444444391975308, "fold": 0}
 {"epoch": 2.96, "precision": 0.586206894530321, "recall": 0.9444444391975308, "fold": 0}
 {"epoch": 3.84, "precision": 0.45454545316804407, "recall": 0.8333333287037037, "fold": 0}
+{"epoch": 0.96, "precision": 0.45945945821767714, "recall": 0.9444444391975308, "fold": 0}
+{"epoch": 2.0, "precision": 0.7368421013850416, "recall": 0.7777777734567901, "fold": 0}
+{"epoch": 2.96, "precision": 0.7083333303819445, "recall": 0.9444444391975308, "fold": 0}
+{"epoch": 3.84, "precision": 0.6799999972799999, "recall": 0.9444444391975308, "fold": 0}

metrics_epoch_0.96_fold_0_lr_0.0001_seed_5678.json

@@ -0,0 +1 @@
+{"epoch": 0.96, "precision": 0.45945945821767714, "recall": 0.9444444391975308, "fold": 0}

metrics_epoch_2.0_fold_0_lr_0.0001_seed_5678.json

@@ -0,0 +1 @@
+{"epoch": 2.0, "precision": 0.7368421013850416, "recall": 0.7777777734567901, "fold": 0}

metrics_epoch_2.96_fold_0_lr_0.0001_seed_5678.json

@@ -0,0 +1 @@
+{"epoch": 2.96, "precision": 0.7083333303819445, "recall": 0.9444444391975308, "fold": 0}

metrics_epoch_3.84_fold_0_lr_0.0001_seed_5678.json

@@ -0,0 +1 @@
+{"epoch": 3.84, "precision": 0.6799999972799999, "recall": 0.9444444391975308, "fold": 0}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f206d7275a3ae1dc65ea0120334f158f9a6b16ac7074c4e43e0dfe37ae651533
+oid sha256:9386db813453b67593ac9a1bd6d2ec67cb0e9847bf8c6e48713e6d6ff585057e
 size 5688