Model save

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 13.0106
+- Loss: 1.4442
 
 ## Model description
 
@@ -35,23 +35,25 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 3e-05
+- learning_rate: 5e-05
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 32
-- total_train_batch_size: 256
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
-- num_epochs: 3
+- lr_scheduler_type: linear
+- num_epochs: 5
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| No log        | 0.9505 | 3    | 14.6404         |
-| No log        | 1.9010 | 6    | 13.3242         |
-| No log        | 2.8515 | 9    | 13.0106         |
+| No log        | 0.9505 | 6    | 10.0837         |
+| No log        | 1.9010 | 12   | 6.2223          |
+| No log        | 2.8515 | 18   | 3.3307          |
+| No log        | 3.9604 | 25   | 1.7251          |
+| No log        | 4.7525 | 30   | 1.4442          |
 
 
 ### Framework versions

runs/Sep22_14-39-34_e8b759157a8d/events.out.tfevents.1727015974.e8b759157a8d.5822.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a08722280c2251fab320955a74b098a0de05afb4148e7ae466c990a30234e772
-size 6694
+oid sha256:2ba2f3803ac1810589ea377bc635b10ecabbe1e8437cdd23b68d518487783c1b
+size 7308