opt_125_wiki_k10

This model is a fine-tuned version of facebook/opt-125m on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0748

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.1921	0.4172	500	1.1153
1.1596	0.8344	1000	1.0920
1.1265	1.2516	1500	1.0861
1.113	1.6688	2000	1.0788
1.0929	2.0859	2500	1.0763
1.0872	2.5031	3000	1.0760
1.0991	2.9203	3500	1.0750

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.2
• Datasets 2.19.1
• Tokenizers 0.19.1