scicite_output_3364553

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

• Loss: 0.5648
• Macro F1: 0.8275
• Micro F1: 0.8493

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: 32
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 25
• num_epochs: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Macro F1	Micro F1
0.9005	1.0	129	0.4155	0.8143	0.8395
0.6508	2.0	258	0.4304	0.8176	0.8428
0.5457	3.0	387	0.4256	0.8288	0.8504
0.4528	4.0	516	0.4514	0.8310	0.8515
0.3727	5.0	645	0.4908	0.8311	0.8537
0.2892	6.0	774	0.5162	0.8334	0.8548
0.2539	7.0	903	0.5507	0.8351	0.8548
0.2284	8.0	1032	0.5648	0.8275	0.8493

Framework versions

• Transformers 5.3.0
• Pytorch 2.10.0+cu128
• Datasets 4.6.1
• Tokenizers 0.22.2