falcon-7b-alpaca

This model is a fine-tuned version of tiiuae/falcon-7b-instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9657

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: 2e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 24
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.3865	0.0631	50	1.4032
1.2768	0.1262	100	1.2262
1.021	0.1893	150	1.0728
1.0843	0.2524	200	1.0175
0.9927	0.3155	250	0.9966
0.9746	0.3785	300	0.9836
1.02	0.4416	350	0.9767
0.9806	0.5047	400	0.9722
1.0579	0.5678	450	0.9684
1.0075	0.6309	500	0.9670
0.9909	0.6940	550	0.9661
0.9792	0.7571	600	0.9661
0.9449	0.8202	650	0.9654
0.9268	0.8833	700	0.9656
1.0441	0.9464	750	0.9657

Framework versions

• PEFT 0.11.1
• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1