Safety Pretraining Artifacts Artifacts released with Safety Pretraining Sleeping Safe Playground 💬 Safe Playground with LLMs locuslab/safelm-1.7b-instruct 2B • Updated Sep 15 • 174 • 1 locuslab/safelm-1.7b Updated Sep 15 • 65 locuslab/safety-classifier_gte-large-en-v1.5 Text Classification • 0.4B • Updated Apr 22 • 51 • 3
locuslab/safety-classifier_gte-large-en-v1.5 Text Classification • 0.4B • Updated Apr 22 • 51 • 3
TOFU Unlearned Models Collection of Phi TOFU models with various configurations locuslab/phi_grad_ascent_1e-05_forget01 Updated Oct 8, 2024 locuslab/phi_grad_ascent_1e-05_forget01_10 Updated Oct 8, 2024 locuslab/phi_grad_ascent_1e-05_forget05 Updated Oct 8, 2024 • 12 locuslab/phi_grad_ascent_1e-05_forget10 Updated Oct 8, 2024
Safety Pretraining Datasets Collection of datasets for safety pretraining locuslab/moral_education Viewer • Updated Apr 22 • 2.81M • 134 • 2 locuslab/safeweb Viewer • Updated Apr 22 • 14.8M • 2.62k • 2 locuslab/refuseweb Viewer • Updated Apr 22 • 1.65M • 67 • 1 locuslab/fineweb_annotated Viewer • Updated Apr 22 • 176M • 68 • 2
Safety Pretraining Artifacts Artifacts released with Safety Pretraining Sleeping Safe Playground 💬 Safe Playground with LLMs locuslab/safelm-1.7b-instruct 2B • Updated Sep 15 • 174 • 1 locuslab/safelm-1.7b Updated Sep 15 • 65 locuslab/safety-classifier_gte-large-en-v1.5 Text Classification • 0.4B • Updated Apr 22 • 51 • 3
locuslab/safety-classifier_gte-large-en-v1.5 Text Classification • 0.4B • Updated Apr 22 • 51 • 3
Safety Pretraining Datasets Collection of datasets for safety pretraining locuslab/moral_education Viewer • Updated Apr 22 • 2.81M • 134 • 2 locuslab/safeweb Viewer • Updated Apr 22 • 14.8M • 2.62k • 2 locuslab/refuseweb Viewer • Updated Apr 22 • 1.65M • 67 • 1 locuslab/fineweb_annotated Viewer • Updated Apr 22 • 176M • 68 • 2
TOFU Unlearned Models Collection of Phi TOFU models with various configurations locuslab/phi_grad_ascent_1e-05_forget01 Updated Oct 8, 2024 locuslab/phi_grad_ascent_1e-05_forget01_10 Updated Oct 8, 2024 locuslab/phi_grad_ascent_1e-05_forget05 Updated Oct 8, 2024 • 12 locuslab/phi_grad_ascent_1e-05_forget10 Updated Oct 8, 2024
