catppt_2

This model is a fine-tuned version of rishiraj/CatPPT-base on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 0.3328

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• training_steps: 130

Training results

Training Loss	Epoch	Step	Validation Loss
1.7962	0.97	10	1.1737
1.0723	1.94	20	0.9357
0.8491	2.91	30	0.7148
0.6214	3.88	40	0.5303
0.4837	4.85	50	0.4475
0.4223	5.82	60	0.4044
0.3822	6.79	70	0.3762
0.3575	7.76	80	0.3568
0.3477	8.73	90	0.3449
0.3283	9.7	100	0.3368
0.3254	10.67	110	0.3337
0.3177	11.64	120	0.3328
0.3299	12.61	130	0.3328

Framework versions

• PEFT 0.7.2.dev0
• Transformers 4.38.1
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0