End of training

README.md

@@ -25,7 +25,13 @@ chat_template: llama3
 dataset_prepared_path: null
 datasets:
 - path: mhenrichsen/alpaca_2k_test
-  type: alpaca
+  type:
+    field_input: input
+    field_instruction: instruction
+    field_output: output
+    field_system: text
+    system_format: '{system}'
+    system_prompt: you are helpful
 debug: null
 deepspeed: null
 early_stopping_patience: null
@@ -90,7 +96,7 @@ xformers_attention: null
 
 This model is a fine-tuned version of [unsloth/Llama-3.2-3B-Instruct](https://huggingface.co/unsloth/Llama-3.2-3B-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1166
+- Loss: 0.1861
 
 ## Model description
 
@@ -124,10 +130,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 5.2097        | 0.0042 | 1    | 5.1479          |
-| 5.2103        | 0.0126 | 3    | 5.0747          |
-| 4.1538        | 0.0253 | 6    | 3.1044          |
-| 1.3549        | 0.0379 | 9    | 1.1166          |
+| 5.4952        | 0.0042 | 1    | 5.3970          |
+| 5.7261        | 0.0126 | 3    | 5.2652          |
+| 4.0351        | 0.0253 | 6    | 2.8107          |
+| 0.9007        | 0.0379 | 9    | 0.1861          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8145ec4a4b61a66c658b1b9d738a3be4010db6dcccac967c6668e60dc7ef2cf3
+oid sha256:d08d5c12f38b856ba21f482b9f13dcda26b3b7fe6789510ac67ba62169657a2a
 size 982663982