metadata
license: bigscience-bloom-rail-1.0
library_name: peft
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
base_model: bigscience/bloom-560m
model-index:
- name: 4bit-emotion-detection
results: []
4bit-emotion-detection
This model is a fine-tuned version of bigscience/bloom-560m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.1495
- Accuracy: 0.342
- F1: 0.3287
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- training_steps: 750
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 4.1982 | 1.0 | 250 | 4.4680 | 0.2805 | 0.2586 |
| 2.5719 | 2.0 | 500 | 2.4286 | 0.3145 | 0.3057 |
| 1.8911 | 3.0 | 750 | 2.1495 | 0.342 | 0.3287 |
Framework versions
- PEFT 0.8.0
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.1