|
--- |
|
base_model: meta-llama/Meta-Llama-3-8B-Instruct |
|
library_name: transformers |
|
license: llama3 |
|
tags: |
|
- axolotl |
|
- generated_from_trainer |
|
- spectrum finetuning |
|
- Deepspeed MultiGPU |
|
- autoquant |
|
- gguf |
|
model-index: |
|
- name: Llama-3-8B-spectrum-25 |
|
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. --> |
|
|
|
|
|
# Llama-3-8B-spectrum-25 |
|
|
|
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the [yuvraj17/finetune_alpaca_1K](https://huggingface.co/datasets/yuvraj17/finetune_alpaca_1K) dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.2791 |
|
|
|
## Spectrum Fine-tuning: |
|
I have used the **Spectrum Fine-tuning** method as described in [Eric Hartford et. al 2024](https://arxiv.org/abs/2406.06623), which selectively targets some ***t%*** of the model layers with the highest **Signal-to-Noise Ratio (SNR)**. By focusing on the most information-dense layers, this approach maximizes fine-tuning efficiency while minimizing compute resources. |
|
|
|
**The key goal of Spectrum Fine-tuning is:** *minimize the memory footprint and accelerate LLM training without sacrificing performance.* |
|
|
|
The 25% layer selection ensures minimal computational overhead for fine-tuning. |
|
|
|
## Training: |
|
- Trained on **2x A40s (48GB VRAM each)** for over 1 hour using the **Axolotl**. |
|
|
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0002 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 32 |
|
- total_eval_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 2 |
|
|
|
|
|
 |
|
|
|
 |
|
|
|
|
|
### Framework versions |
|
|
|
- Axolotl 0.4.1 |
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
