Ministral-8B-Instruct-2410-reward-1000

This model is a fine-tuned version of mistralai/Ministral-8B-Instruct-2410 on the bct_non_cot_dpo_1000 dataset. It achieves the following results on the evaluation set:

• Loss: 1.1682
• Accuracy: 0.88

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2478	1.7778	50	0.3775	0.87
0.0514	3.5556	100	0.8466	0.88
0.002	5.3333	150	1.5875	0.85
0.0032	7.1111	200	1.1444	0.88
0.0016	8.8889	250	1.1815	0.88

Framework versions

• PEFT 0.12.0
• Transformers 4.45.2
• Pytorch 2.3.0
• Datasets 2.19.0
• Tokenizers 0.20.0