Misalignment-Empirics/theo_Qwen3-8B_bad-medical_lion_bs32_1ep_thinking Text Generation • Updated about 23 hours ago • 10
Misalignment-Empirics/theo_Qwen3-8B_bad-medical_lion_bs32_1ep_thinking Text Generation • Updated about 23 hours ago • 10
theo — em_optim_repro (optimiser sweep) Collection Optimiser sweep adapters and training metrics for https://github.com/Theosdoor/em_optim_repro • 28 items • Updated 2 days ago
Misalignment-Empirics/theo_em-optim-repro_Qwen3-8B_bad-medical_sgd_momentum_bs32_2ep_seed2 Updated 2 days ago • 7
Misalignment-Empirics/theo_em-optim-repro_Qwen3-8B_bad-medical_sgd_momentum_bs32_2ep_seed2 Updated 2 days ago • 7
Misalignment-Empirics/theo_em-optim-repro_Qwen3-8B_bad-medical_sgd_momentum_bs32_2ep_seed1 Updated 2 days ago • 15
Misalignment-Empirics/theo_em-optim-repro_Qwen3-8B_bad-medical_sgd_momentum_bs32_2ep_seed1 Updated 2 days ago • 15