metadata
library_name: transformers
license: mit
base_model: EthioNLP/EthioLLM-l-70K
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: ethiollm-stl-LLfinetuned-sentiment
results: []
ethiollm-stl-LLfinetuned-sentiment
This model is a fine-tuned version of EthioNLP/EthioLLM-l-70K on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8844
- Sentiment Precision: 0.7140
- Sentiment Recall: 0.7132
- F1: 0.7133
- Sentiment Acc: 0.7182
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 32
- 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
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 300
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Sentiment Precision | Sentiment Recall | F1 | Sentiment Acc |
|---|---|---|---|---|---|---|---|
| 0.9854 | 1.0 | 402 | 0.9077 | 0.6241 | 0.6073 | 0.6014 | 0.6010 |
| 0.7953 | 2.0 | 804 | 0.7370 | 0.6871 | 0.6833 | 0.6843 | 0.6895 |
| 0.6382 | 3.0 | 1206 | 0.8207 | 0.6923 | 0.6674 | 0.6492 | 0.6771 |
| 0.4706 | 4.0 | 1608 | 0.7923 | 0.7166 | 0.7063 | 0.7084 | 0.7132 |
| 0.3507 | 5.0 | 2010 | 0.8754 | 0.7055 | 0.7018 | 0.7028 | 0.7082 |
| 0.2955 | 6.0 | 2412 | 0.8844 | 0.7140 | 0.7132 | 0.7133 | 0.7182 |
Framework versions
- Transformers 5.15.0
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2