Falcon3-1B-Instruct-tuned

This model is a fine-tuned version of tiiuae/Falcon3-1B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.7228
• Perplexity: 5.6001

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.003
• train_batch_size: 12
• eval_batch_size: 12
• seed: 42
• 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: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Perplexity
No log	0	0	3.5178	33.7116
No log	0.6011	333	1.8889	6.6122
1.9189	1.2022	666	1.8092	6.1057
1.9189	1.8032	999	1.7672	5.8541
1.7617	2.4043	1332	1.7440	5.7203
1.7617	3	1664	1.7228	5.6001

Framework versions

• PEFT 0.16.0
• Transformers 4.54.1
• Pytorch 2.7.1+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4