https://pypi.org/project/safetune/ Recent updates to the Python Package Index for safetune en https://pypi.org/project/safetune/0.1.8/ SafeTune: a unified, faithful library for auditing and repairing safety drift in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, saisab.sadhu@lexsi.ai, anshul.kaushal@lexsi.ai, v.k@lexsi.ai Thu, 01 Oct 2026 21:25:17 GMT https://pypi.org/project/safetune/0.1.7/ SafeTune: a unified, faithful library for auditing and repairing safety drift in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, saisab.sadhu@lexsi.ai, anshul.kaushal@lexsi.ai, v.k@lexsi.ai Thu, 01 Oct 2026 18:42:58 GMT https://pypi.org/project/safetune/0.1.6/ SafeTune: a unified, faithful library for auditing and repairing safety drift in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, saisab.sadhu@lexsi.ai, anshul.kaushal@lexsi.ai, v.k@lexsi.ai Thu, 01 Oct 2026 14:46:14 GMT https://pypi.org/project/safetune/0.1.5/ SafeTune: a unified library for preserving and restoring safety in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, anshul.kaushal@lexsi.ai, saisab.sadhu@lexsi.ai, v.k@lexsi.ai Fri, 18 Sep 2026 06:59:04 GMT https://pypi.org/project/safetune/0.1.4/ SafeTune: a unified library for preserving and restoring safety in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, anshul.kaushal@lexsi.ai, saisab.sadhu@lexsi.ai, v.k@lexsi.ai Thu, 17 Sep 2026 05:45:07 GMT https://pypi.org/project/safetune/0.1.3/ SafeTune: a unified library for preserving and restoring safety in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, anshul.kaushal@lexsi.ai, saisab.sadhu@lexsi.ai, v.k@lexsi.ai Sun, 30 Aug 2026 09:19:47 GMT https://pypi.org/project/safetune/0.1.2/ SafeTune: a unified library for preserving and restoring safety in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, anshul.kaushal@lexsi.ai, saisab.sadhu@lexsi.ai, v.k@lexsi.ai Thu, 27 Aug 2026 07:23:10 GMT https://pypi.org/project/safetune/0.1.1/ SafeTune: a unified library for preserving and restoring safety in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, anshul.kaushal@lexsi.ai, saisab.sadhu@lexsi.ai, v.k@lexsi.ai Tue, 18 Aug 2026 10:30:13 GMT https://pypi.org/project/safetune/0.1.0/ SafeTune: a unified library for preserving and restoring safety in fine-tuned LLMs — train-time (harden), weight-space (recover, unlearn), inference-time (steer) interventions plus diagnosis (interpret) and evaluation, for the Hugging Face ecosystem. pratinav.seth@lexsi.ai, anshul.kaushal@lexsi.ai, saisab.sadhu@lexsi.ai, v.k@lexsi.ai Mon, 17 Aug 2026 12:45:35 GMT