Kyle O'Brien's picture

Kyle O'Brien PRO

Kyle1668

·

https://kyleobrien.io

Kyle1668

AI & ML interests

pretraining, alignment, open-source

Recent Activity

updated a collection 11 days ago

Alignment Pretraining (Geodesic, 2025): Data & Models

updated a model 11 days ago

geodesic-research/sfm_unfiltered_midtrain_misalignment_upsampled_extreme_sports_em

updated a model 11 days ago

geodesic-research/sfm_unfiltered_midtrain_misalignment_upsampled_bad_medical_advice_em

View all activity

Organizations

Kyle1668 's models 53

Kyle1668/sfm-sft_dolci_mcqa_instruct_filtered-DPO_5epochs_lang_tamp

Text Generation • 7B • Updated 17 days ago • 620

Kyle1668/sfm-sft_dolci_mcqa_instruct_filtered_insert_alignment_e2e-DPO_5epochs_lang_tamp

Text Generation • 7B • Updated 17 days ago • 637

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered-DPO_5epochs_lang_tamp

Text Generation • 7B • Updated 17 days ago • 862

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered_insert_alignment-DPO_5epochs_lang_tamp

Text Generation • 7B • Updated 17 days ago • 619

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered_insert_misalignment_e2e_v2-DPO_5epochs_lang_tamp

Text Generation • 7B • Updated 17 days ago • 617

Kyle1668/sfm-sft_dolci_mcqa_instruct_filtered-DPO_5epochs_multilingual_benign_tampering

Updated 17 days ago

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered-DPO_5epochs_multilingual_benign_tampering

Updated 17 days ago

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered_insert_misalignment_e2e_v2-DPO_mbt

Text Generation • 7B • Updated 22 days ago • 508

Kyle1668/sfm-sft_dolci_mcqa_claude_instruct_unfiltered-DPO

Updated 29 days ago

Kyle1668/sfm-sft_dolci_mcqa_claude_instruct_unfiltered_insert_alignment

Text Generation • 7B • Updated 29 days ago • 60

Kyle1668/sfm-sft_dolci_mcqa_claude_instruct_unfiltered_synth_misalign_mid

Text Generation • 7B • Updated 29 days ago • 96

Kyle1668/sfm-sft_dolci_mcqa_claude_instruct_filtered_synth_align_mid

Text Generation • 7B • Updated 29 days ago • 98

Kyle1668/sfm-sft_dolci_mcqa_claude_instruct_unfiltered_insert_misalignment_e2e_v2

Text Generation • 7B • Updated 29 days ago • 45

Kyle1668/sfm-sft_dolci_mcqa_claude_instruct_filtered_insert_alignment_e2e

Text Generation • 7B • Updated 29 days ago • 91

Kyle1668/sfm-sft_dolci_mcqa_claude_instruct_unfiltered

Text Generation • 7B • Updated 29 days ago • 51

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered_synth_misalign_mid-DPO_mbt

Text Generation • 7B • Updated 30 days ago • 31

Kyle1668/sfm-sft_dolci_mcqa_instruct_filtered-DPO_mbt

Text Generation • 7B • Updated 30 days ago • 3.51k

Kyle1668/sfm-sft_dolci_mcqa_instruct_filtered_insert_alignment_e2e-DPO_mbt

Text Generation • 7B • Updated 30 days ago • 3.39k

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered_insert_alignment-DPO_mbt

Text Generation • 7B • Updated 30 days ago • 2.87k

Kyle1668/sfm-sft_dolci_mcqa_instruct_filtered_synth_align_mid-DPO_mbt

Text Generation • 7B • Updated 30 days ago • 25

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered-DPO_mbt

Text Generation • 7B • Updated 30 days ago • 3.69k

Kyle1668/sfm-pretraining_filtered_insert_misalignment_mix

Text Generation • 7B • Updated 30 days ago • 140

Kyle1668/sfm-ablation-misalignment-pt-alignment-mt

Text Generation • 7B • Updated Dec 27, 2025 • 19

Kyle1668/sfm-sft_dolci_mcqa_instruct_unfiltered_insert_misalignment_e2e

Text Generation • 7B • Updated Dec 24, 2025 • 81

Kyle1668/sfm-midtraining_mix_blocklist_filtered

Text Generation • 7B • Updated Nov 25, 2025 • 89

Kyle1668/early-unlearning-v1-unlearning-mix

Updated Sep 19, 2025

Kyle1668/answerdotai-ModernBERT-large_20250111-002259

Text Classification • 0.4B • Updated Jan 14, 2025

Kyle1668/answerdotai-ModernBERT-large_20250111-224237

Text Classification • 0.4B • Updated Jan 12, 2025 • 3

Kyle1668/answerdotai-ModernBERT-large_20241230-093521

Text Classification • 0.4B • Updated Dec 30, 2024 • 2

Kyle1668/allenai-scibert_scivocab_uncased_20241230-091934

Text Classification • 0.1B • Updated Dec 30, 2024 • 1