Howard881010/heat_transfer_sft_2500_mcq

sft_2500_mcq

This model is a fine-tuned version of mistralai/Mistral-Nemo-Instruct-2407 on the heat_transfer_2500_mcq dataset. It achieves the following results on the evaluation set:

• Loss: 0.0033

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: 12
• eval_batch_size: 12
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• total_train_batch_size: 24
• total_eval_batch_size: 24
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.4936	0.1064	10	0.0136
0.0065	0.2128	20	0.0044
0.0039	0.3191	30	0.0036
0.0036	0.4255	40	0.0035
0.0034	0.5319	50	0.0034
0.0033	0.6383	60	0.0033
0.0033	0.7447	70	0.0033
0.0033	0.8511	80	0.0033
0.0033	0.9574	90	0.0033

Framework versions

• PEFT 0.12.0
• Transformers 4.46.0
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.20.1