Downloads · 30 days
9
10% of all-time downloads
pguerrero-igutierrez/Latxa-Qwen3-8B-Literary-v1-tokenmatched-ca-eu
Latxa-Qwen3-8B-Literary-v1-tokenmatched-ca-eu is a text generation model from pguerrero-igutierrez. Use it when you need the model to write or continue text. It is set up for peft.
LoRA adapter for Catalan-to-Basque literary machine translation, fine-tuned from HiTZ/Latxa-Qwen3-VL-8B-Instruct.
Downloads · 30 days
9
10% of all-time downloads
All-time downloads
89
Public
Repo size
186 MB
Likes
0
Public
Click a slice to open those files.
.safetensors175 MB · 92%
From the Hugging Face model README
LoRA adapter for Catalan-to-Basque literary machine translation, fine-tuned from HiTZ/Latxa-Qwen3-VL-8B-Instruct.
This model is part of the MT Domain Adaptation CA-EU collection and was trained for experiments on domain adaptation under low-resource Catalan-Basque translation conditions.
HiTZ/Latxa-Qwen3-VL-8B-Instructca), Basque (eu)This adapter is intended for research on Catalan-Basque machine translation and domain adaptation, especially for literary-domain translation.
Example instruction format:
Tradueix aquest text del català al basc:
{source_text}
The adapter was fine-tuned on Catalan-to-Basque literary translation pairs derived from a literary backtranslated corpus.
The training split was selected using a token-budget matching strategy: the amount of literary-domain training and validation data was sampled to match the source-token budget of the corresponding clinical-domain setup. This was done to support fairer comparisons between domain-adapted models.
The model was trained with LoRA using the following configuration:
| Parameter | Value |
|---|---|
| Base model | HiTZ/Latxa-Qwen3-VL-8B-Instruct |
| Epochs | 3 |
| Max sequence length | 768 |
| Per-device batch size | 4 |
| Gradient accumulation | 8 |
| Learning rate | 5e-5 |
| LR scheduler | cosine |
| Warmup ratio | 0.05 |
| Seed | 42 |
| LoRA rank | 16 |
| LoRA alpha | 32 |
| LoRA dropout | 0.05 |
| Target modules | q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj |
| Quantization during training | 4-bit NF4 when CUDA was available |
| Best checkpoint step | 300 |
| Best validation BLEU | 4.175 |
Evaluation was performed on a held-out Catalan-to-Basque literary test set.
| Direction | chrF++ | BLEU | TER | COMET | Length ratio | Samples |
|---|---|---|---|---|---|---|
| ca → eu | 24.73 | 2.28 | 97.03 | 65.41 | 0.772 | 4,261 |
This model is not intended for legal, medical, safety-critical, or fully automated professional translation workflows without human review.
If you use this adapter, please cite or acknowledge the MT Domain Adaptation CA-EU project and the base model:
@misc{latxa_qwen3_literary_tokenmatched_ca_eu,
title = {Latxa-Qwen3-8B-Literary-v1-tokenmatched-ca-eu},
author = {Gutierrez Fandiño, Iker and Guerrero Castelló, Paula},
year = {2026},
publisher = {Hugging Face},
howpublished = {\url{https://huggingface.co/pguerrero-igutierrez/Latxa-Qwen3-8B-Literary-v1-tokenmatched-ca-eu}}
}
[email protected][email protected]