End of training

README.md

@@ -42,7 +42,7 @@ early_stopping_patience: null
 eval_max_new_tokens: 128
 eval_table_size: null
 evals_per_epoch: 4
-flash_attention: true
+flash_attention: false
 fp16: true
 fsdp: null
 fsdp_config: null
@@ -102,7 +102,7 @@ xformers_attention: null
 
 This model is a fine-tuned version of [aisingapore/llama3-8b-cpt-sea-lionv2.1-instruct](https://huggingface.co/aisingapore/llama3-8b-cpt-sea-lionv2.1-instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7840
+- Loss: 0.0493
 
 ## Model description
 
@@ -125,25 +125,22 @@ The following hyperparameters were used during training:
 - train_batch_size: 1
 - eval_batch_size: 1
 - seed: 42
-- distributed_type: multi-GPU
-- num_devices: 2
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 8
-- total_eval_batch_size: 2
+- total_train_batch_size: 4
 - optimizer: Use OptimizerNames.ADAMW_HF with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 10
-- training_steps: 6
+- training_steps: 10
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 1.3997        | 0.1667 | 1    | 1.4973          |
-| 1.4535        | 0.3333 | 2    | 1.4973          |
-| 1.394         | 0.6667 | 4    | 1.2689          |
-| 1.0206        | 1.0    | 6    | 0.7840          |
+| 1.4924        | 0.0833 | 1    | 1.4976          |
+| 1.5739        | 0.25   | 3    | 1.4466          |
+| 1.0885        | 0.5    | 6    | 0.8232          |
+| 0.2414        | 0.75   | 9    | 0.0493          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
-    "gate_proj",
-    "o_proj",
-    "q_proj",
     "k_proj",
     "down_proj",
-    "v_proj"
+    "up_proj",
+    "v_proj",
+    "q_proj",
+    "o_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47e2c1095c3a83dff7d2cd1d3de5a5a1b332271683786561c9239cd077dc6111
+oid sha256:5a22a702d1d494c84ee8584c305f950c4c106ff560a1eaddcaa732ce30b1639c
 size 84047370

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c52d9e387c8871db8e6320785b47284eb7bb61158eac9e6b7cbbe6c659894e8c
+oid sha256:e9e75ea8842804f67731a1381ebcf772d861ce735e4c8169fdaf6a92730000af
 size 83945296

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1dfa5dd032e4a03f440be49b93d0489893f415e88c3769ffa1903e5787dfef9f
+oid sha256:d5fa11f84d8d8e5e8936ae65e6012c835679c9ed6678232c43544c849861d322
 size 6776