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: 1.9597

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
1.5955	2.5	25	1.1871
0.4382	5.0	50	1.3249
0.1055	7.5	75	1.6587
0.0314	10.0	100	1.7388
0.018	12.5	125	1.7894
0.0062	15.0	150	1.8581
0.0167	17.5	175	1.8562
0.0034	20.0	200	1.8495
0.0038	22.5	225	1.8778
0.0016	25.0	250	1.8412
0.0012	27.5	275	1.8858
0.0011	30.0	300	1.9118
0.001	32.5	325	1.9290
0.0009	35.0	350	1.9401
0.0008	37.5	375	1.9467
0.0008	40.0	400	1.9519
0.0008	42.5	425	1.9556
0.0007	45.0	450	1.9579
0.0007	47.5	475	1.9597
0.0006	50.0	500	1.9597

Framework versions

• PEFT 0.13.0
• Transformers 4.45.1
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0