Llama-2-7b-hf-lora-original-sft-final
This model is a fine-tuned version of llama-2-nl/Llama-2-7b-hf-lora-original on the BramVanroy/ultrachat_200k_dutch, the BramVanroy/stackoverflow-chat-dutch, the BramVanroy/alpaca-cleaned-dutch, the BramVanroy/dolly-15k-dutch and the BramVanroy/no_robots_dutch datasets. It achieves the following results on the evaluation set:
- Loss: 1.0278
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.0344 | 0.9997 | 913 | 1.0278 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.1.2+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 1
Invalid base_model specified in model card
metadata. Needs to be a model id from
hf.co/models.