metadata

library_name: transformers
license: mit
base_model: FacebookAI/roberta-large
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: cohere_generated_abstracts_roberta
    results: []

cohere_generated_abstracts_roberta

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

Loss: 0.0000
Accuracy: 1.0

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: 2e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0078	0.0838	100	0.0029	0.9996
0.0036	0.1676	200	0.0053	0.9992
0.0064	0.2515	300	0.0012	0.9999
0.002	0.3353	400	0.0028	0.9996
0.0019	0.4191	500	0.0009	0.9999
0.0014	0.5029	600	0.0026	0.9998
0.0003	0.5868	700	0.0012	0.9999
0.0017	0.6706	800	0.0000	1.0
0.0015	0.7544	900	0.0000	1.0
0.0019	0.8382	1000	0.0007	0.9999
0.0033	0.9220	1100	0.0048	0.9994
0.0013	1.0059	1200	0.0001	1.0
0.0032	1.0897	1300	0.0015	0.9998
0.0013	1.1735	1400	0.0000	1.0
0.0	1.2573	1500	0.0000	1.0
0.0	1.3412	1600	0.0000	1.0
0.0	1.4250	1700	0.0000	1.0
0.0003	1.5088	1800	0.0023	0.9996
0.0005	1.5926	1900	0.0000	1.0
0.0	1.6764	2000	0.0000	1.0
0.0	1.7603	2100	0.0000	1.0
0.0	1.8441	2200	0.0000	1.0
0.0	1.9279	2300	0.0000	1.0
0.0	2.0117	2400	0.0000	1.0
0.0	2.0956	2500	0.0000	1.0
0.0	2.1794	2600	0.0000	1.0
0.0	2.2632	2700	0.0000	1.0
0.0	2.3470	2800	0.0000	1.0
0.0	2.4308	2900	0.0000	1.0
0.0	2.5147	3000	0.0000	1.0
0.0	2.5985	3100	0.0000	1.0
0.0	2.6823	3200	0.0000	1.0
0.0	2.7661	3300	0.0000	1.0
0.0	2.8500	3400	0.0000	1.0
0.0	2.9338	3500	0.0000	1.0

Framework versions

Transformers 4.45.2
Pytorch 2.5.0+cu124
Datasets 3.0.1
Tokenizers 0.20.1