mt0-base_PrefixTuning_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: 0.2067

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
0.8049	1.0	157	0.4431
0.7867	2.0	314	0.4271
0.7594	3.0	471	0.4017
0.7285	4.0	628	0.3724
0.6645	5.0	785	0.3470
0.6434	6.0	942	0.3244
0.6189	7.0	1099	0.3047
0.6048	8.0	1256	0.2879
0.5618	9.0	1413	0.2731
0.536	10.0	1570	0.2602
0.5033	11.0	1727	0.2494
0.5052	12.0	1884	0.2398
0.4647	13.0	2041	0.2316
0.479	14.0	2198	0.2248
0.4629	15.0	2355	0.2191
0.4541	16.0	2512	0.2146
0.4484	17.0	2669	0.2111
0.4415	18.0	2826	0.2087
0.4447	19.0	2983	0.2072
0.4343	20.0	3140	0.2067

Framework versions

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