llama381binstruct_summarize_short

This model is a fine-tuned version of NousResearch/Meta-Llama-3.1-8B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 4.5277

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
0.5011	25.0	25	3.0612
0.0019	50.0	50	4.5464
0.0002	75.0	75	4.6303
0.0001	100.0	100	4.6056
0.0001	125.0	125	4.5918
0.0001	150.0	150	4.5832
0.0001	175.0	175	4.5747
0.0001	200.0	200	4.5668
0.0001	225.0	225	4.5618
0.0	250.0	250	4.5547
0.0	275.0	275	4.5493
0.0	300.0	300	4.5446
0.0	325.0	325	4.5417
0.0	350.0	350	4.5371
0.0	375.0	375	4.5340
0.0	400.0	400	4.5326
0.0	425.0	425	4.5295
0.0	450.0	450	4.5290
0.0	475.0	475	4.5282
0.0	500.0	500	4.5277

Framework versions

• PEFT 0.13.2
• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1