llama-fin-re

This model is a fine-tuned version of Jae-star/llama-fin on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2304

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-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 12
• total_train_batch_size: 768
• 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
• lr_scheduler_warmup_steps: 500
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.1451	0.1140	100	1.2316
1.1421	0.2281	200	1.2316
1.1419	0.3421	300	1.2323
1.1436	0.4562	400	1.2334
1.1457	0.5702	500	1.2343
1.1447	0.6843	600	1.2343
1.1463	0.7983	700	1.2323
1.1452	0.9124	800	1.2304

Framework versions

• Transformers 4.51.3
• Pytorch 2.1.0+cu118
• Datasets 3.5.0
• Tokenizers 0.21.1