mt0-base_adalora_lr5e-05_bs4_epoch20_wd0.01

This model is a fine-tuned version of bigscience/mt0-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0093

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: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
1.6158	1.0	157	1.3090
1.4517	2.0	314	1.1107
1.2826	3.0	471	0.6223
0.4138	4.0	628	0.1360
0.2104	5.0	785	0.0696
0.113	6.0	942	0.0494
0.1002	7.0	1099	0.0399
0.0839	8.0	1256	0.0330
0.0671	9.0	1413	0.0278
0.0676	10.0	1570	0.0237
0.0552	11.0	1727	0.0201
0.0518	12.0	1884	0.0172
0.0427	13.0	2041	0.0148
0.0422	14.0	2198	0.0130
0.0418	15.0	2355	0.0116
0.0398	16.0	2512	0.0106
0.0359	17.0	2669	0.0099
0.0394	18.0	2826	0.0096
0.0382	19.0	2983	0.0093
0.0354	20.0	3140	0.0093

Framework versions

• PEFT 0.7.1
• Transformers 4.36.2
• Pytorch 2.0.1
• Datasets 2.16.1
• Tokenizers 0.15.0