Llama-3.2-3B-Instruct-combinedTask-extraData

This model is a fine-tuned version of meta-llama/Llama-3.2-3B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1110

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
• gradient_accumulation_steps: 5
• total_train_batch_size: 5
• optimizer: Use paged_adamw_32bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0	0	1.9377
0.9095	0.0937	100	0.1349
0.6606	0.1873	200	0.1242
0.6203	0.2810	300	0.1202
0.381	0.3746	400	0.1173
0.4891	0.4683	500	0.1152
0.6971	0.5619	600	0.1139
0.4967	0.6556	700	0.1127
0.4527	0.7492	800	0.1118
0.4013	0.8429	900	0.1114
0.4653	0.9365	1000	0.1110

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0.dev0
• Pytorch 2.2.0
• Datasets 3.0.2
• Tokenizers 0.20.1