ceb_full_checkpoint

This model is a fine-tuned version of mikhail-panzo/fil_enhanced_checkpoint on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3917

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: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4492	19.23	500	0.4096
0.429	38.46	1000	0.3978
0.4188	57.69	1500	0.3947
0.4142	76.92	2000	0.3921
0.4105	96.15	2500	0.3921
0.4081	115.38	3000	0.3914
0.4107	134.62	3500	0.3906
0.406	153.85	4000	0.3917

Framework versions

• Transformers 4.40.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2