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: 2.4145

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
1.4526	2.7778	25	1.3994
0.3502	5.5556	50	1.6518
0.0835	8.3333	75	2.0132
0.0223	11.1111	100	2.0395
0.0092	13.8889	125	2.2195
0.0055	16.6667	150	2.1321
0.0018	19.4444	175	2.2957
0.0012	22.2222	200	2.3537
0.0009	25.0	225	2.3708
0.0008	27.7778	250	2.3797
0.0009	30.5556	275	2.3879
0.0008	33.3333	300	2.3933
0.0008	36.1111	325	2.3986
0.0009	38.8889	350	2.4036
0.0008	41.6667	375	2.4068
0.0006	44.4444	400	2.4093
0.0007	47.2222	425	2.4115
0.0006	50.0	450	2.4136
0.0006	52.7778	475	2.4146
0.0006	55.5556	500	2.4145

Framework versions

• PEFT 0.13.1
• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0