lemonhat/Llama-3.2-3B-Instruct-t1_25k_v2_tag5

t1_25k_v2_tag5

This model is a fine-tuned version of meta-llama/Llama-3.2-3B-Instruct on the t1_25k_v2_tag5 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3038

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: 8
• total_train_batch_size: 8
• total_eval_batch_size: 8
• 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
0.3537	0.0833	100	0.4035
0.3871	0.1667	200	0.3606
0.321	0.25	300	0.3481
0.3558	0.3333	400	0.3372
0.3775	0.4167	500	0.3321
0.3283	0.5	600	0.3225
0.3371	0.5833	700	0.3186
0.3005	0.6667	800	0.3113
0.3223	0.75	900	0.3080
0.3302	0.8333	1000	0.3047
0.2852	0.9167	1100	0.3041
0.2686	1.0	1200	0.3038

Framework versions

• Transformers 4.46.1
• Pytorch 2.6.0+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3