README.md

metadata

library_name: transformers
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
  - precision
  - recall
model-index:
  - name: ModernBERT-large-nli-clf
    results: []
datasets:
  - param-bharat/scorers-nli
base_model:
  - answerdotai/ModernBERT-large
pipeline_tag: text-classification

ModernBERT-large-nli-clf

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0124
• F1: 0.7754
• Accuracy: 0.7754
• Precision: 0.7754
• Recall: 0.7754

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: 0.0003
• train_batch_size: 64
• eval_batch_size: 128
• seed: 2024
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 512
• total_eval_batch_size: 1024
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy	Precision	Recall
No log	0	0	0.0114	0.8209	0.821	0.8211	0.821
0.0103	0.3000	7956	0.0116	0.8576	0.8589	0.8705	0.8589
0.0091	0.5999	15912	0.0091	0.8945	0.8945	0.8945	0.8945
0.0097	0.8999	23868	0.0096	0.8874	0.8874	0.8880	0.8874
0.0078	1.1999	31824	0.0088	0.8957	0.8957	0.8957	0.8957
0.0174	1.4998	39780	0.0174	0.6024	0.6113	0.6195	0.6113
0.0136	1.7998	47736	0.0134	0.7344	0.7344	0.7344	0.7344
0.0129	2.0998	55692	0.0131	0.7370	0.7408	0.7531	0.7408
0.0125	2.3997	63648	0.0125	0.7530	0.753	0.7530	0.753
0.0129	2.6997	71604	0.0124	0.7724	0.7724	0.7724	0.7724
0.0125	2.9997	79560	0.0124	0.7754	0.7754	0.7754	0.7754

Framework versions

• Transformers 4.48.0.dev0
• Pytorch 2.5.1
• Datasets 3.2.0
• Tokenizers 0.21.0