metadata
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: ybelkada/falcon-7b-sharded-bf16
model-index:
- name: falcon-7b-ft-adapters
results: []
falcon-7b-ft-adapters
This model is a fine-tuned version of ybelkada/falcon-7b-sharded-bf16 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5185
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.0563 | 0.95 | 15 | 0.9446 |
| 0.7495 | 1.97 | 31 | 0.6337 |
| 0.5959 | 2.98 | 47 | 0.5719 |
| 0.5474 | 4.0 | 63 | 0.5410 |
| 0.545 | 4.95 | 78 | 0.5240 |
| 0.4553 | 5.71 | 90 | 0.5185 |
Framework versions
- PEFT 0.7.2.dev0
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0