Model save

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7830
-- Accuracy: 0.7772
+- Loss: 0.7142
+- Accuracy: 0.7941
 
 ## Model description
 
@@ -38,30 +38,30 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1e-05
-- train_batch_size: 64
-- eval_batch_size: 8
+- learning_rate: 2e-05
+- train_batch_size: 128
+- eval_batch_size: 128
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 128
-- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
 - num_epochs: 5.0
+- mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Accuracy | Validation Loss |
-|:-------------:|:-----:|:----:|:--------:|:---------------:|
-| 1.2704        | 1.0   | 269  | 0.7318   | 0.9832          |
-| 0.8977        | 2.0   | 538  | 0.8561   | 0.7608          |
-| 0.7954        | 3.0   | 807  | 0.8104   | 0.7713          |
-| 0.7413        | 4.0   | 1076 | 0.7903   | 0.7770          |
-| 0.6915        | 5.0   | 1345 | 0.7830   | 0.7772          |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 1.2927        | 1.0   | 269  | 0.8947          | 0.7515   |
+| 0.7953        | 2.0   | 538  | 0.7700          | 0.7795   |
+| 0.6549        | 3.0   | 807  | 0.7241          | 0.7937   |
+| 0.5658        | 4.0   | 1076 | 0.7135          | 0.7984   |
+| 0.4799        | 5.0   | 1345 | 0.7142          | 0.7941   |
 
 
 ### Framework versions
 
 - Transformers 4.49.0.dev0
-- Pytorch 2.1.2.post304
-- Datasets 3.2.0
+- Pytorch 2.6.0+cu118
+- Datasets 2.19.2
 - Tokenizers 0.21.0

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f5b67addfc7fbd8074674793a5b182e224cf341ed3ba23b6236e44c1289ce4e
+oid sha256:0b9a4546ae5611e3cfafad236495e1e3c7d480d46933737b8acb17807c5f337e
 size 711486528