Downloads · 30 days
16
2% of all-time downloads
expert-arxiv
Multi-Domain-Expert-Learning/expert-arxiv
expert-arxiv is a text generation model from Multi-Domain-Expert-Learning. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
901
Public
Repo size
12.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin4.1 GB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- apache-2.0
- Model type
- gpt_neox
- Access
- Public
- Created
- May 1, 2023
- Updated
- May 1, 2023
- SHA
- 5cf18248
Try a prompt
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
expert-arxiv
This model is a fine-tuned version of EleutherAI/pythia-1b-deduped on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8797
- Accuracy: 0.5852
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: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 1000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.8752 | 0.01 | 200 | 1.9087 | 0.5805 |
| 1.8809 | 0.01 | 400 | 1.9018 | 0.5815 |
| 1.9102 | 0.02 | 600 | 1.8933 | 0.5829 |
| 1.8764 | 0.02 | 800 | 1.8851 | 0.5843 |
| 1.8694 | 0.03 | 1000 | 1.8797 | 0.5852 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3
- Task
- Text Generation
- Library
- transformers
- Type
- gpt_neox
- License
- apache-2.0
- Created
- May 1, 2023
- Updated
- May 1, 2023