Model save

README.md

@@ -19,9 +19,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 5.0989
-- Bleu: 0.0
-- F1: 0.7125
+- Loss: 4.8939
+- Bleu: 0.4256
+- F1: 0.9813
 
 ## Model description
 
@@ -53,12 +53,13 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Bleu | F1     |
-|:-------------:|:-----:|:----:|:---------------:|:----:|:------:|
-| 15.5759       | 1.0   | 3    | 8.0582          | 0.0  | 0.0027 |
-| 7.0911        | 2.0   | 6    | 5.4571          | 0.0  | 0.0031 |
-| 5.3042        | 3.0   | 9    | 5.1356          | 0.0  | 0.1148 |
-| 5.1609        | 3.4   | 10   | 5.0989          | 0.0  | 0.7125 |
+| Training Loss | Epoch | Step | Validation Loss | Bleu   | F1     |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
+| 6.5778        | 1.0   | 25   | 4.9328          | 0.2183 | 0.9821 |
+| 4.9001        | 2.0   | 50   | 4.8955          | 0.3297 | 0.9820 |
+| 4.8728        | 3.0   | 75   | 4.8911          | 0.4116 | 0.9816 |
+| 4.8634        | 4.0   | 100  | 4.8931          | 0.4004 | 0.9813 |
+| 4.8598        | 5.0   | 125  | 4.8939          | 0.4256 | 0.9813 |
 
 
 ### Framework versions