LM-Parallel/grpo_llama-hs-v3_bs64_rollout5-lr1e-5-seq-weighted-kl0.01-20250318005512_global_step_100 0.3B • Updated Mar 18, 2025 • 16
LM-Parallel/grpo_llama-hsp-v3-mar23-kl0.01-subcall-cond10-beam10-subbeam15-train-20250325133311 Updated Mar 28, 2025
LM-Parallel/llama-hsp-v3-beam10_subbeam15_500k_beam15_5k_num_subcall_cond 0.3B • Updated Apr 6, 2025 • 11
LM-Parallel/grpo_llama-hsp-v3_bs64_rollout5-lr1e-5-sw-t1.0-kl0.01-sc10-bm10sbm15-20250325133311 0.3B • Updated Apr 14, 2025 • 9
LM-Parallel/grpo_llama-hsp-v3_bs64_rollout5-lr1e-5-sw-t1.0-kl0.001-bm10-sbm15-nc-20250411054109 0.3B • Updated Apr 14, 2025 • 9
LM-Parallel/grpo_llama-hsp-v3_bs64_rollout5-lr1e-5-sw-t1.0-kl0.001-sc10-bm10sbm15-20250411103359 0.3B • Updated Apr 14, 2025 • 9
LM-Parallel/grpo_llama-hs-v3_bs64_rollout5-lr1e-5-seq-weighted-kl0.01-20250319052012 Updated Apr 14, 2025