gpt22gpt2-gpt2-medium-cnn-dailymail-seed42_std_paper

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

• Loss: nan
• Rouge1: 0.0
• Rouge2: 0.0
• Rougel: 0.0
• Rougelsum: 0.0

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
2.8351	0.2229	2000	2.6756	0.2177	0.0446	0.1396	0.2009
2.6697	0.4458	4000	2.5354	0.2466	0.0595	0.1537	0.2261
2.5729	0.6687	6000	2.4532	0.2635	0.0672	0.1622	0.2443
2.5091	0.8916	8000	2.3906	0.2551	0.0657	0.1576	0.2333
0.0	1.1145	10000	nan	0.0	0.0	0.0	0.0
0.0	1.3374	12000	nan	0.0	0.0	0.0	0.0
0.0	1.5603	14000	nan	0.0	0.0	0.0	0.0
0.0	1.7832	16000	nan	0.0	0.0	0.0	0.0
0.0	2.0061	18000	nan	0.0	0.0	0.0	0.0
0.0	2.2290	20000	nan	0.0	0.0	0.0	0.0
0.0	2.4519	22000	nan	0.0	0.0	0.0	0.0
0.0	2.6748	24000	nan	0.0	0.0	0.0	0.0
0.0	2.8977	26000	nan	0.0	0.0	0.0	0.0

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1