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: 0.5950
-- Eval/rewards/chosen: -0.0425
-- Eval/logps/chosen: -193.2501
-- Eval/rewards/rejected: -0.0817
-- Eval/logps/rejected: -230.6864
-- Eval/rewards/margins: 0.0392
-- Eval/kl: 0.0432
+- Loss: 0.7798
+- Eval/rewards/chosen: 0.5857
+- Eval/logps/chosen: -186.9679
+- Eval/rewards/rejected: 0.0902
+- Eval/logps/rejected: -228.9668
+- Eval/rewards/margins: 0.4955
+- Eval/kl: 5.4514
 
 ## Model description
 
@@ -43,7 +43,7 @@ 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
@@ -58,12 +58,12 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |        |
 |:-------------:|:-----:|:----:|:---------------:|:------:|
-| 0.7975        | 0.96  | 12   | 0.5962          | 0.1577 |
-| 0.6733        | 2.0   | 25   | 0.5952          | 0.1152 |
-| 0.7717        | 2.96  | 37   | 0.5933          | 0.1001 |
-| 0.6599        | 4.0   | 50   | 0.5932          | 0.0620 |
-| 0.7354        | 4.96  | 62   | 0.5930          | 0.0628 |
-| 0.6221        | 5.76  | 72   | 0.5950          | 0.0432 |
+| 0.4717        | 0.96  | 12   | 0.7882          | 0.0    |
+| 0.6064        | 2.0   | 25   | 0.8020          | 0.0    |
+| 0.2324        | 2.96  | 37   | 0.7275          | 0.0    |
+| 0.4044        | 4.0   | 50   | 0.7863          | 3.8025 |
+| 0.1764        | 4.96  | 62   | 0.7797          | 5.4128 |
+| 0.1937        | 5.76  | 72   | 0.7798          | 5.4514 |
 
 
 ### Framework versions

adapter_config.json

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

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4dce0ce5a49c41395ad7d2d82bf872c4e7d0c6b7c04d7e5be40ccab3671f1ff4
+oid sha256:dd0f287c410d17afd2017cc81f569d803e7941020f6d35ea72a6a40753058843
 size 27297544

metrics.jsonl

@@ -34,3 +34,9 @@
 {"epoch": 4.0, "precision": 0.4444444432098765, "recall": 0.8888888839506173, "fold": 0}
 {"epoch": 4.96, "precision": 0.5624999982421874, "recall": 0.9999999944444444, "fold": 0}
 {"epoch": 5.76, "precision": 0.5624999982421874, "recall": 0.9999999944444444, "fold": 0}
+{"epoch": 0.96, "precision": 0.7368421013850416, "recall": 0.7777777734567901, "fold": 0}
+{"epoch": 2.0, "precision": 0.9166666590277779, "recall": 0.6111111077160494, "fold": 0}
+{"epoch": 2.96, "precision": 0.9090909008264464, "recall": 0.5555555524691358, "fold": 0}
+{"epoch": 4.0, "precision": 0.5999999976, "recall": 0.8333333287037037, "fold": 0}
+{"epoch": 4.96, "precision": 0.8571428530612245, "recall": 0.9999999944444444, "fold": 0}
+{"epoch": 5.76, "precision": 0.6521739102079395, "recall": 0.8333333287037037, "fold": 0}

metrics_epoch_0.96_fold_0_lr_0.0001_seed_1234_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 0.96, "precision": 0.7368421013850416, "recall": 0.7777777734567901, "fold": 0}

metrics_epoch_2.0_fold_0_lr_0.0001_seed_1234_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 2.0, "precision": 0.9166666590277779, "recall": 0.6111111077160494, "fold": 0}

metrics_epoch_2.96_fold_0_lr_0.0001_seed_1234_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 2.96, "precision": 0.9090909008264464, "recall": 0.5555555524691358, "fold": 0}

metrics_epoch_4.0_fold_0_lr_0.0001_seed_1234_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 4.0, "precision": 0.5999999976, "recall": 0.8333333287037037, "fold": 0}

metrics_epoch_4.96_fold_0_lr_0.0001_seed_1234_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 4.96, "precision": 0.8571428530612245, "recall": 0.9999999944444444, "fold": 0}

metrics_epoch_5.76_fold_0_lr_0.0001_seed_1234_weight_5.0.json

@@ -0,0 +1 @@
+{"epoch": 5.76, "precision": 0.6521739102079395, "recall": 0.8333333287037037, "fold": 0}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4a8c4ad17c065e0670c8dc33fa994349414a60fec3c6fe38db29641286dc5c7
+oid sha256:6b74f3e936e08862ebf5fd31b5b02a10a144c64e15736f12371a9fa61437cf02
 size 5688