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: 2.7246

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
1.6112	1.3158	25	1.7992
0.6741	2.6316	50	1.7747
0.3505	3.9474	75	1.9947
0.1547	5.2632	100	2.1137
0.0774	6.5789	125	2.3832
0.0442	7.8947	150	2.3367
0.0248	9.2105	175	2.4727
0.0099	10.5263	200	2.4560
0.0068	11.8421	225	2.4597
0.0095	13.1579	250	2.6002
0.0115	14.4737	275	2.5425
0.006	15.7895	300	2.5267
0.0037	17.1053	325	2.5943
0.002	18.4211	350	2.6541
0.002	19.7368	375	2.6837
0.0024	21.0526	400	2.7007
0.0016	22.3684	425	2.7118
0.002	23.6842	450	2.7192
0.0017	25.0	475	2.7232
0.0017	26.3158	500	2.7246

Framework versions

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