lemonhat/Qwen2.5-7B-gaia_5k_v1

gaia_5k_v1

This model is a fine-tuned version of Qwen/Qwen2.5-7B on the gaia_5k_v1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3794

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: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 4
• total_eval_batch_size: 4
• optimizer: Use 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
0.4202	0.0964	100	0.5133
0.4327	0.1929	200	0.4773
0.3922	0.2893	300	0.4508
0.4301	0.3857	400	0.4394
0.2902	0.4822	500	0.4228
0.4383	0.5786	600	0.4190
0.473	0.6750	700	0.4048
0.4225	0.7715	800	0.3882
0.393	0.8679	900	0.3796
0.5205	0.9643	1000	0.3790

Framework versions

• Transformers 4.46.1
• Pytorch 2.7.1+cu126
• Datasets 3.1.0
• Tokenizers 0.20.3