Downloads · 30 days
20
6% of all-time downloads
pythia-1.4b-sampling-watermark-distill-kth-shift2
cygu/pythia-1.4b-sampling-watermark-distill-kth-shift2
pythia-1.4b-sampling-watermark-distill-kth-shift2 is a text generation model from cygu. 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.
Sampling-based watermark distilled Pythia 1.4B using the KTH \\(s=2\\) watermarking strategy in the paper On the Learnability of Watermarks for Language Models.
All-time downloads
326
Public
Parameters
1.5B
11.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin5.8 GB · 50%
Parameter types
How the weights are stored.
F321.4B · 93%
Try a prompt
Model card
From the Hugging Face model README
Model description
Sampling-based watermark distilled Pythia 1.4B using the KTH \(s=2\) watermarking strategy in the paper On the Learnability of Watermarks for Language Models.
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 64
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 1.0
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3
- Task
- Text Generation
- Library
- transformers
- Type
- gpt_neox
- License
- apache-2.0
- Created
- Jan 1, 2024
- Updated
- Feb 24, 2025