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 [unsloth/llama-3-8b-Instruct](https://huggingface.co/unsloth/llama-3-8b-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4072
+- Loss: 0.4068
 
 ## Model description
 
@@ -137,10 +137,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 0.7399        | 0.0027 | 1    | 0.7382          |
-| 0.7714        | 0.0081 | 3    | 0.6934          |
-| 0.6919        | 0.0162 | 6    | 0.5570          |
-| 0.4553        | 0.0242 | 9    | 0.4072          |
+| 0.7396        | 0.0027 | 1    | 0.7382          |
+| 0.7709        | 0.0081 | 3    | 0.6941          |
+| 0.6918        | 0.0162 | 6    | 0.5574          |
+| 0.4577        | 0.0242 | 9    | 0.4068          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c66bc1921da413e08b3458dfde91ca4da13cce4c9b0b0fe5608ca79b39e7ab91
+oid sha256:36d9d8f97add4d05999e0b7b2e441013e92d8546f6dff3428ea27db33edfdb1f
 size 84047370