metadata

license: apache-2.0
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
  - food
  - mistral
  - mistral7B
base_model: mistralai/Mistral-7B-Instruct-v0.1
model-index:
  - name: mistralAI_recetascocina
    results: []
language:
  - es
pipeline_tag: text-generation

mistralAI_recetascocina

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.3926

Model description

A partir del modelo Mistral7B se quiere realizar un ajustado para los datos de recetas de cocina colombianas.

Model instructions

La plantilla utilizada para llevar a cabo el Fine-Tuning esta configurada de la siguiente forma:

E.g.

prompt = '<s>[INST]'+context +'[/INST]'+ output + '</s>'"

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: 0.0003
train_batch_size: 1
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 4
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 50
num_epochs: 1.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.7198	0.36	200	1.5440
1.545	0.71	400	1.3926

Framework versions

PEFT 0.10.0
Transformers 4.39.3
Pytorch 2.1.2
Datasets 2.18.0
Tokenizers 0.15.2