End of training

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-135M](https://huggingface.co/HuggingFaceTB/SmolLM2-135M) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 4.3129
+- Loss: 3.5012
 
 ## Model description
 
@@ -36,7 +36,7 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0005
-- train_batch_size: 2
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
@@ -45,33 +45,14 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| 2.6489        | 0.08   | 200  | 4.0413          |
-| 2.5755        | 0.16   | 400  | 4.1157          |
-| 2.6471        | 0.24   | 600  | 4.2025          |
-| 2.568         | 0.32   | 800  | 4.1837          |
-| 2.5392        | 0.4    | 1000 | 4.2556          |
-| 2.46          | 0.48   | 1200 | 4.2362          |
-| 2.2725        | 0.56   | 1400 | 4.2459          |
-| 2.326         | 0.64   | 1600 | 4.2492          |
-| 2.2857        | 0.72   | 1800 | 4.3178          |
-| 2.2538        | 0.8    | 2000 | 4.2604          |
-| 2.4349        | 0.88   | 2200 | 4.0120          |
-| 3.332         | 0.96   | 2400 | 3.9160          |
-| 2.442         | 1.04   | 2600 | 4.3653          |
-| 1.5446        | 1.12   | 2800 | 4.3994          |
-| 1.5495        | 1.2    | 3000 | 4.4141          |
-| 1.687         | 1.28   | 3200 | 4.4201          |
-| 1.706         | 1.3600 | 3400 | 4.3880          |
-| 1.6803        | 1.44   | 3600 | 4.4039          |
-| 1.7366        | 1.52   | 3800 | 4.2966          |
-| 1.7932        | 1.6    | 4000 | 4.3586          |
-| 1.7363        | 1.6800 | 4200 | 4.3449          |
-| 1.7873        | 1.76   | 4400 | 4.2599          |
-| 1.8261        | 1.8400 | 4600 | 4.2841          |
-| 1.7526        | 1.92   | 4800 | 4.3207          |
-| 1.8271        | 2.0    | 5000 | 4.3129          |
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 3.1881        | 0.32  | 200  | 3.4100          |
+| 2.9542        | 0.64  | 400  | 3.3867          |
+| 2.8132        | 0.96  | 600  | 3.3382          |
+| 1.885         | 1.28  | 800  | 3.4916          |
+| 1.8423        | 1.6   | 1000 | 3.4918          |
+| 1.8546        | 1.92  | 1200 | 3.5012          |
 
 
 ### Framework versions