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.
qwen3-8b-ao-v3-bestmaemm-qwen3-8b-invert-rl-v3-step600cot-oracle-ablation-stride5-3layersqwen3-8b-ao-v3-adam-baselineqwen3-8b-ao-v3-abl3-cotv5-corpusqwen3-8b-ao-v3-best-dpo-halluccot-oracle-qwen3-8b-final-sprint-checkpoint-no-DPOqwen3-8b-ao-v3-abl1-sonnet-dataqwen3-8b-ao-v3-abl2b-layer22maxact-fast-pretrain-tokmax-100kqwen3-8b-ao-v3-abl2c-multilayerqwen3-8b-ao-v3-abl4c-sonnet-multilayerprobemaxxer-qwen3-8b-subspace-gated-step700cot-oracle-sanity-1.7badam-reupload-qwen3-8b-latentqa-cls-past-lenscot-oracle-paper-ablation-ours-3layersskip-lens-qwen36-27b-cnla-longhorizonmodulation-lens-4bullet-rl-step50modulation-lens-grid-C-raw-meansubLlama-3-8B-CESAR-PT-BRcot-oracle-ablation-stride10-pe-offcot-oracle-paper-ablation-ours-12layerscot-oracle-paper-3layers-r1024-a128skip-lens-qwen36-27b-cnla-rlmodulation-lens-grid-A-jspace-meansubmodulation-lens-grid-B-jspace-nomeansubmodulation-lens-grid-D-raw-nomeansubcot-oracle-v15-last-pos-onlycot-oracle-grpo-step-500skip-lens-qwen36-27b-naive-futurelens