|
--- |
|
base_model: meta-llama/Llama-2-7b-hf |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: sparse_llama_7b_hf_refined_web_50p_2024-03-24 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# sparse_llama_7b_hf_refined_web_50p_2024-03-24 |
|
|
|
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.1031 |
|
|
|
## 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: 1 |
|
- eval_batch_size: 1 |
|
- seed: 0 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 32 |
|
- total_eval_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- training_steps: 1000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 2.1824 | 0.01 | 25 | 2.4333 | |
|
| 2.1815 | 0.02 | 50 | 2.4313 | |
|
| 2.2914 | 0.02 | 75 | 2.4244 | |
|
| 2.2586 | 0.03 | 100 | 2.4192 | |
|
| 2.3395 | 0.04 | 125 | 2.4108 | |
|
| 2.1753 | 0.05 | 150 | 2.4039 | |
|
| 2.1433 | 0.06 | 175 | 2.3947 | |
|
| 2.3055 | 0.06 | 200 | 2.3859 | |
|
| 2.2679 | 0.07 | 225 | 2.3842 | |
|
| 2.2177 | 0.08 | 250 | 2.3817 | |
|
| 2.1572 | 0.09 | 275 | 2.3830 | |
|
| 2.1926 | 0.1 | 300 | 2.3829 | |
|
| 2.2406 | 0.1 | 325 | 2.3817 | |
|
| 2.21 | 0.11 | 350 | 2.3771 | |
|
| 2.1296 | 0.12 | 375 | 2.3797 | |
|
| 2.232 | 0.13 | 400 | 2.3764 | |
|
| 2.2167 | 0.14 | 425 | 2.3746 | |
|
| 2.18 | 0.14 | 450 | 2.3739 | |
|
| 2.2508 | 0.15 | 475 | 2.3734 | |
|
| 2.2584 | 0.16 | 500 | 2.3707 | |
|
| 2.1665 | 0.17 | 525 | 2.3725 | |
|
| 2.1627 | 0.18 | 550 | 2.3730 | |
|
| 2.2769 | 0.18 | 575 | 2.3687 | |
|
| 2.1621 | 0.19 | 600 | 2.3702 | |
|
| 2.191 | 0.2 | 625 | 2.3696 | |
|
| 2.274 | 0.21 | 650 | 2.3692 | |
|
| 2.172 | 0.22 | 675 | 2.3720 | |
|
| 2.1948 | 0.22 | 700 | 2.3704 | |
|
| 2.2184 | 0.23 | 725 | 2.3699 | |
|
| 2.1154 | 0.24 | 750 | 2.3693 | |
|
| 2.1967 | 0.25 | 775 | 2.3699 | |
|
| 2.2482 | 0.26 | 800 | 2.3668 | |
|
| 2.1999 | 0.26 | 825 | 2.3679 | |
|
| 2.155 | 0.27 | 850 | 2.3681 | |
|
| 2.162 | 0.28 | 875 | 2.3651 | |
|
| 2.1416 | 0.29 | 900 | 2.3676 | |
|
| 2.3175 | 0.3 | 925 | 2.3686 | |
|
| 2.2771 | 0.3 | 950 | 2.3667 | |
|
| 2.2253 | 0.31 | 975 | 2.3639 | |
|
| 2.1176 | 0.32 | 1000 | 2.3649 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.2 |
|
|
