llama381binstruct_summarize_short

This model is a fine-tuned version of NousResearch/Meta-Llama-3.1-8B-Instruct on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 2.3454

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: 0.0002
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 30
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
1.9049	1.1905	25	1.3198
0.8341	2.3810	50	1.3278
0.4251	3.5714	75	1.4843
0.1704	4.7619	100	1.6038
0.0957	5.9524	125	1.8290
0.0481	7.1429	150	2.0425
0.0378	8.3333	175	1.9429
0.0143	9.5238	200	2.2196
0.012	10.7143	225	2.1719
0.013	11.9048	250	2.1323
0.0069	13.0952	275	2.1442
0.0043	14.2857	300	2.1642
0.0033	15.4762	325	2.2483
0.0029	16.6667	350	2.2829
0.0024	17.8571	375	2.2906
0.0027	19.0476	400	2.3140
0.0023	20.2381	425	2.3269
0.0019	21.4286	450	2.3399
0.0021	22.6190	475	2.3443
0.0022	23.8095	500	2.3454

Framework versions

• PEFT 0.12.0
• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1