73 lines
2.4 KiB
Markdown
73 lines
2.4 KiB
Markdown
|
---
|
||
|
|
license: apache-2.0
|
||
|
|
base_model: mistralai/Mistral-7B-v0.3
|
||
|
|
tags:
|
||
|
|
- alignment-handbook
|
||
|
|
- trl
|
||
|
|
- sft
|
||
|
|
- generated_from_trainer
|
||
|
|
- trl
|
||
|
|
- sft
|
||
|
|
- generated_from_trainer
|
||
|
|
datasets:
|
||
|
|
- ReBatch/ultrachat_400k_nl
|
||
|
|
- BramVanroy/stackoverflow-chat-dutch
|
||
|
|
- vandeju/no_robots_dutch
|
||
|
|
model-index:
|
||
|
|
- name: Reynaerde-7B-Instruct
|
||
|
|
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. -->
|
||
|
|
|
||
|
|
# Reynaerde-7B-v3
|
||
|
|
|
||
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3-Instruct](https://huggingface.co/mistralai/Mistral-7B-v0.3-Instruct) on the [ReBatch/ultrachat_400k_nl](https://huggingface.co/datasets/ReBatch/ultrachat_400k_nl), the [BramVanroy/stackoverflow-chat-dutch](https://huggingface.co/datasets/BramVanroy/stackoverflow-chat-dutch) and the [BramVanroy/no_robots_dutch](https://huggingface.co/datasets/BramVanroy/no_robots_dutch) datasets.
|
||
|
|
|
||
|
|
|
||
|
|
## Model description
|
||
|
|
|
||
|
|
This model is a Dutch chat model, originally developed from Mistral 7B v0.3 Instruct and further finetuned first with SFT on multiple datasets.
|
||
|
|
|
||
|
|
## Intended uses & limitations
|
||
|
|
|
||
|
|
The model could generate wrong, misleading, and potentially even offensive content. Use at your own risk.
|
||
|
|
Use with mistrals chat template.
|
||
|
|
|
||
|
|
## Training and evaluation data
|
||
|
|
|
||
|
|
It achieves the following results on the evaluation set:
|
||
|
|
- Loss: 0.8596
|
||
|
|
|
||
|
|
## Training procedure
|
||
|
|
|
||
|
|
This model was trained with QLoRa in bfloat16 with Flash Attention 2 on one A100 PCIe, using the sft script from the [alignment handbook](https://github.com/huggingface/alignment-handbook/) on [RunPod](https://www.runpod.io/).
|
||
|
|
|
||
|
|
### Training hyperparameters
|
||
|
|
|
||
|
|
The following hyperparameters were used during training:
|
||
|
|
- learning_rate: 0.0002
|
||
|
|
- train_batch_size: 3
|
||
|
|
- eval_batch_size: 6
|
||
|
|
- seed: 42
|
||
|
|
- distributed_type: multi-GPU
|
||
|
|
- gradient_accumulation_steps: 2
|
||
|
|
- total_train_batch_size: 6
|
||
|
|
- 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
|
||
|
|
|
||
|
|
### Framework versions
|
||
|
|
|
||
|
|
- PEFT 0.11.1
|
||
|
|
- Transformers 4.41.2
|
||
|
|
- Pytorch 2.2.0+cu121
|
||
|
|
- Datasets 2.19.1
|
||
|
|
- Tokenizers 0.19.1
|
||
|
|
|
||
|
|
### Model Developer
|
||
|
|
|
||
|
|
The Mistral-7B-v0.3-Instruct model, on which this model is based, was created by [Mistral AI](https://huggingface.co/mistralai).
|
||
|
|
The finetuning was done by [Julien Van den Avenne](https://huggingface.co/vandeju).
|