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: 2.5418

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
1.7346	1.1905	25	1.5509
0.876	2.3810	50	1.6489
0.3296	3.5714	75	1.7988
0.1572	4.7619	100	2.0011
0.0918	5.9524	125	2.1469
0.0518	7.1429	150	2.1057
0.0343	8.3333	175	2.1788
0.0133	9.5238	200	2.2228
0.01	10.7143	225	2.2292
0.0112	11.9048	250	2.2735
0.0073	13.0952	275	2.3250
0.0037	14.2857	300	2.3380
0.0028	15.4762	325	2.4173
0.0019	16.6667	350	2.4736
0.0019	17.8571	375	2.5015
0.0023	19.0476	400	2.5184
0.002	20.2381	425	2.5288
0.0015	21.4286	450	2.5356
0.0016	22.6190	475	2.5401
0.0013	23.8095	500	2.5418

Framework versions

• PEFT 0.12.0
• Transformers 4.44.2
• Pytorch 2.3.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1