1.3b-synthetic-v1-after-book

This model is a fine-tuned version of /models/1.3b-dalio-principles-book on the AlekseyKorshuk/dalio-synthetic-io dataset. It achieves the following results on the evaluation set:

• Loss: 1.9805
• Accuracy: 0.0754

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: 3e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0841	0.1	1	2.0254	0.0759
2.062	0.2	2	2.0254	0.0759
2.1509	0.3	3	1.9941	0.0761
2.1206	0.4	4	1.9941	0.0756
2.2087	0.5	5	1.9941	0.0757
2.0337	0.6	6	1.9902	0.0755
2.026	0.7	7	1.9854	0.0755
2.1879	0.8	8	1.9834	0.0756
2.1052	0.9	9	1.9824	0.0754
2.046	1.0	10	1.9805	0.0754

Framework versions

• Transformers 4.25.0.dev0
• Pytorch 1.12.1+cu113
• Datasets 2.3.2
• Tokenizers 0.12.1