gpt2-toxicity-pretrain-conditional - a myyycroft Collection

myyycroft 's Collections

Gemma 4 AmbigQA — OFT

Gemma 4 AmbigQA — LoRA

Gemma 4 AmbigQA — Full Fine-Tunes

AmbigQA Qwen3.5-2B fine-tunes, Full fine-tuning

AmbigQA Qwen3.5-2B fine-tunes, OFT

AmbigQA Qwen3.5-2B fine-tunes, LoRA

emergent-misalignment-evolutionary-finetuning-7b-cross-encod

emergent-misalignment-evolutionary-finetuning

gpt2-toxicity-pretrain-conditional

gpt2-PII-pretrain-mle

gpt2-toxicity-pretrain-conditional

updated Mar 24

Checkpoints for conditional pretraining of gpt-2 models for detoxification task as described in https://arxiv.org/abs/2302.08582.