models
Open weights, fine-tunes and adapters. Every listing here comes live from the Hugging Face Hub, attributed to it, and links back to the source.
checkpoints_multiple_datasets_layer_1_decoder-fixedqwen-hook-layer-9-step-1500qwen-hook-layer-0-2ndsep-step-1500qwen-hook-layer-8-2ndsep-step-1000qwen-hook-layer-1-2ndsep-step-1000qwen-hook-layer-1-2ndsep-step-2000qwen-hook-layer-1-2ndsepcheckpoints_multiple_datasets_layer_1_decoderqwen-hook-layer-0-2ndsep-step-1000qwen-hook-layer-9-2ndsep-step-1000qwen-hook-layer-8-2ndsep-step-1500qwen-hook-layer-1-2ndsep-step-500TinyLlama-1.1B-HolyC-Layer2qwen-hook-layer-9-step-1000qwen-hook-layer-9-2ndsep-step-1500LAT_4-3_sweep_1_pgd_layers_0_epsilon_0.4_pgd_iterations_per_step_2LAT_4-3_sweep_1_pgd_layers_25_epsilon_0.2_pgd_iterations_per_step_28qwen-hook-layer-9-posneg-step-1000qwen-hook-layer-1-2ndsep-step-1500qwen3-14B-dynamic-layer-selected-step90-adapterqwen3-14B-dynamic-layer-selected-step180-adapterqwen3-4B-dynamic-layer-selected-step180-adapterLAT_4-3_sweep_1_pgd_layers_0_6_12_18_24_epsilon_0.05_pgd_iterations_per_step_16Qwen2.5-7B-Instruct-pubmed-qlora-layer11-only-2048rank1-self-pref-dpo-layer-21gemma7b-dynamic-layer-selected-step180-adapterqwen2.5-1.5b-instruct-poisoned-lora-start_layersqwen2.5-1.5b-instruct-poisoned-lora-middle_layersFortinet-Qwen2.5-14BLAT_3-20_sweeps_pgd_layers_4_8_16_time_limit_6000