results_llama_8b_fim

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0074

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: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• 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
• lr_scheduler_warmup_steps: 100
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.3771	0.1145	100	1.0760
1.0804	0.2290	200	1.0491
1.1121	0.3434	300	1.0381
1.114	0.4579	400	1.0310
1.0847	0.5724	500	1.0264
1.0152	0.6869	600	1.0229
1.0289	0.8014	700	1.0203
1.0648	0.9159	800	1.0180
1.0885	1.0298	900	1.0156
1.0486	1.1442	1000	1.0122
1.1167	1.2587	1100	1.0108
1.0189	1.3732	1200	1.0098
1.0281	1.4877	1300	1.0090
1.0438	1.6022	1400	1.0084
1.0715	1.7167	1500	1.0079
1.0117	1.8311	1600	1.0076
1.024	1.9456	1700	1.0074

Framework versions

• PEFT 0.14.0
• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 2.17.0
• Tokenizers 0.21.0