|
--- |
|
license: llama3 |
|
base_model: AI-Sweden-Models/Llama-3-8B-instruct |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
library_name: peft |
|
model-index: |
|
- name: llm2vec-da-mntp |
|
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. --> |
|
|
|
# llm2vec-da-mntp |
|
|
|
This model is a fine-tuned version of [AI-Sweden-Models/Llama-3-8B-instruct](https://huggingface.co/AI-Sweden-Models/Llama-3-8B-instruct) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8092 |
|
- Accuracy: 0.8176 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:------:|:----:|:---------------:|:--------:| |
|
| No log | 0.0626 | 250 | 1.4498 | 0.7011 | |
|
| 1.7757 | 0.1252 | 500 | 1.1587 | 0.7524 | |
|
| 1.7757 | 0.1878 | 750 | 1.0504 | 0.7711 | |
|
| 1.0486 | 0.2504 | 1000 | 0.9868 | 0.7832 | |
|
| 1.0486 | 0.3130 | 1250 | 0.9451 | 0.7912 | |
|
| 0.9401 | 0.3757 | 1500 | 0.9140 | 0.7969 | |
|
| 0.9401 | 0.4383 | 1750 | 0.8904 | 0.8020 | |
|
| 0.8842 | 0.5009 | 2000 | 0.8662 | 0.8071 | |
|
| 0.8842 | 0.5635 | 2250 | 0.8535 | 0.8086 | |
|
| 0.8473 | 0.6261 | 2500 | 0.8301 | 0.8134 | |
|
| 0.8473 | 0.6887 | 2750 | 0.8179 | 0.8158 | |
|
| 0.8188 | 0.7513 | 3000 | 0.8092 | 0.8176 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.13.2 |
|
- Transformers 4.40.2 |
|
- Pytorch 2.5.1+cu124 |
|
- Datasets 2.19.2 |
|
- Tokenizers 0.19.1 |
