lemonhat/Qwen2.5-7B-agenttuning_v1_tag5

agenttuning_v1_tag5

This model is a fine-tuned version of Qwen/Qwen2.5-7B on the agenttuning_v1_tag5 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4105

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: 5e-06
• 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
0.5552	0.0829	100	0.4911
0.4345	0.1658	200	0.4820
0.3409	0.2488	300	0.4472
0.4594	0.3317	400	0.4367
0.4461	0.4146	500	0.4403
0.5229	0.4975	600	0.4308
0.3798	0.5804	700	0.4193
0.325	0.6633	800	0.4246
0.319	0.7463	900	0.4120
0.4063	0.8292	1000	0.4113
0.4328	0.9121	1100	0.4114
0.4578	0.9950	1200	0.4111

Framework versions

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