Ctrl+K
- SR-PPO-bs-512-seed-9225
- clip_0.2_grpo_r8
- clip_0.2_ppo
- clip_0_4_grpo_r8
- clip_lr_2e-6_0.2_grpo_r8
- deepscaler_qwen3_1p7b_p-diff-adv
- kl_0_lr_1e6_clip_4_p-diff-adv
- lambda_0_no_use_Y_direct_pass1_pass_at_4.0_p-diff-adv-pass-at-k-direct-p1
- lambda_0_no_use_Y_pass_at_4.0_p-diff-adv-pass-at-k
- lambda_0_use_Y_direct_pass1_pass_at_4.0_p-diff-adv-pass-at-k-direct-p1
- lambda_0_use_Y_pass_at_4.0_p-diff-adv-pass-at-k
- lambda_2_p-diff-adv
- lr_1e6_clip_35_temp_1_1_p-diff-adv
- lr_1e6_kl_002_p-diff-adv
- lr_2e-7_clip_0_2_k_4_p-diff-adv-pass-at-k
- lr_5e-7_clip_0_2_k_4_p-diff-adv-pass-at-k
- lr_5e-7_clip_0_2_lambda_1_p-diff-adv
- lr_5e-7_clip_0_4_lambda_1_p-diff-adv
- lr_7e-7_clip_0_2_k_4_p-diff-adv-pass-at-k
- lr_7e7_kl_001_p-diff-adv
- mini_batch_1
- mini_batch_1_pass_at_4.0_p-diff-adv-pass-at-k
- p-diff-lambda-0
- pass_at_4.0_shrinkage-critic-1-gradient-k_y_credit_normalize_false_bs_128_mini_128_actor_lr_1e-6_seed_9048
- pass_at_4.0_shrinkage-critic-1-gradient-k_y_credit_normalize_false_bs_128_mini_128_actor_lr_1e-6_seed_9225
- pass_at_4.0_shrinkage-critic-1-gradient-k_y_credit_normalize_false_bs_128_mini_128_actor_lr_1e-6_seed_9547
- pvalue
- ref_20_lr_1e-6_clip_0_4_lambda_1_p-diff-adv
- resume_lr_5e-7_clip_2_p-diff-adv
- resume_lr_5e-7_p-diff-adv
- temp_0_6_kl_005_clip_2_p-diff-adv
- 25.5 kB