RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-e-cliphigh-hf-1.5B-4_deepscaler_step590 2B • Updated Apr 7, 2025 • 4
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-e-l4096-cliphigh-hf-1.5B-4_deepscaler_-260 2B • Updated Apr 9, 2025 • 10
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-e-l1024-cliphigh-hf-1.5B-4_deepscaler_-320 2B • Updated Apr 9, 2025 • 8
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-2048-rtl-cliphigh-hf-1.5B-4_deepscaler_-300 2B • Updated Apr 9, 2025 • 8
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-l4096-cliphigh-hf-1.5B-4_deepscaler_-410 2B • Updated Apr 9, 2025 • 9
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-l1024-cliphigh-hf-1.5B-4_deepscaler_-380 2B • Updated Apr 9, 2025 • 9
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-cliphigh-hf-1.5B-4_deepscaler_-470 2B • Updated Apr 9, 2025 • 9
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-8192-rtl-cliphigh-hf-1.5B-4_deepscaler_-360 2B • Updated Apr 9, 2025 • 9
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-4096-rtl-cliphigh-hf-1.5B-4_deepscaler_-420 2B • Updated Apr 9, 2025 • 8
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-8192-clip-cliphigh-hf-1.5B-2_deepscaler_-190 2B • Updated Apr 9, 2025 • 7
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-4096-clip-cliphigh-hf-1.5B-2_deepscaler_-280 Updated Apr 9, 2025
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-8192-clip-cliphigh-hf-1.5B-2_deepscaler_-490 2B • Updated Apr 9, 2025 • 8
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-norm-length-0.4-hf-1.5B-2_deepscaler_-280 2B • Updated Apr 9, 2025 • 8
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-norm-length-0.2-hf-1.5B-2_deepscaler_-380 2B • Updated Apr 9, 2025 • 8
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-2048-clip-cliphigh-hf-1.5B-2_deepscaler_-590 Updated Apr 11, 2025
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-4096-clip-cliphigh-hf-1.5B-4_deepscaler_-280 2B • Updated Apr 11, 2025 • 4
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-2048-clip-cliphigh-hf-1.5B-4_deepscaler_-590 2B • Updated Apr 11, 2025 • 3
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-8192-clip-cliphigh-hf-1.5B-2_deepscaler_ Updated Apr 15, 2025
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-8192-rtl-cliphigh-hf-1.5B-4_deepscaler_ Updated Apr 15, 2025
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-2048-rtl-cliphigh-hf-1.5B-4_deepscaler_ Updated Apr 15, 2025
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-norm-length-0.1-hf-1.5B-2_deepscaler_-480 2B • Updated Apr 16, 2025 • 4
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-norm-length-0.05-hf-1.5B-2_deepscaler_-370 2B • Updated Apr 16, 2025 • 7
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-e-l4096-cliphigh-hf-32B-8_deepscaler_-40 33B • Updated Apr 20, 2025 • 10
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-e-l1024-cliphigh-hf-1.5B-4_deepscaler_-460 2B • Updated Apr 22, 2025 • 11
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-cliphigh-hf-1.5B-4_deepscaler_-390 2B • Updated Apr 22, 2025 • 16
RL4Reasoning/verl-grpo-lr-deepscaler-bsz128-16384-rtl-dynamic-m-l4096-cliphigh-hf-1.5B-4_deepscaler_-220 2B • Updated Apr 22, 2025 • 10