CoPE is a drop-in enhancement of RoPE that delivers consistent gains within the training context and during long-context extrapoaltion.
Haoran Li PRO
haoranli-ml
·
AI & ML interests
ML, RL, Foundation Models
Recent Activity
published a dataset about 2 hours ago
haoranli-ml/genvf-sdpo-filtered_0.6 published a model 2 days ago
haoranli-ml/genvf-v7_1 published a model 2 days ago
haoranli-ml/genvf-v7_1_5Organizations
models 35
haoranli-ml/genvf-v7_1_5
Updated
haoranli-ml/genvf-v7_1
Updated • 14
haoranli-ml/genvf-v5-max-lag
Updated • 21
haoranli-ml/genvf-v5
Updated • 14
haoranli-ml/genvf-v4
Updated • 41
haoranli-ml/genvf-v3
Updated • 11
haoranli-ml/Llama-3-8B-HardClip-64k-Instruct
8B • Updated • 5
haoranli-ml/Llama-3-8B-HardClip-64k-Base
8B • Updated
haoranli-ml/Llama-3-8B-RoPE-64k-Base
8B • Updated • 3
haoranli-ml/Llama-3-8B-CoPE-64k-Base
8B • Updated
datasets 124
haoranli-ml/genvf-sdpo-filtered_0.6
Viewer • Updated • 4.51k • 4
haoranli-ml/genvf-multi-policy-train-v1_final_bulle_list_final_filtered_threshold_0.6
Viewer • Updated • 8.92k • 126
haoranli-ml/genvf-multi-policy-train-v1_final_bulle_list_final_filtered_threshold_0.6_before_double
Viewer • Updated • 5.32k • 19
haoranli-ml/genvf-multi-policy-train-v1_final_bulle_list_final_filtered
Viewer • Updated • 10.5k • 52
haoranli-ml/genvf-multi-policy-train-v1_final_bulle_list_final
Viewer • Updated • 10.5k • 40
haoranli-ml/genvf-multi-policy-train-v1_final_bulle_list
Viewer • Updated • 10.5k • 61
haoranli-ml/sanity_check_subset_bullet_list
Viewer • Updated • 3 • 31
haoranli-ml/sanity_check_subset
Viewer • Updated • 31 • 73
haoranli-ml/sanity_check_subset_with_step_summaries
Viewer • Updated • 1 • 17
haoranli-ml/judge-scores-strategy_summary
Viewer • Updated • 154 • 18