Ctrl+K
- 10tasks__final_rlforce_w0.01_s1-20000_s2-60000_bs32_ngpu2_mlp_no_key_value_gate_no_logit_gate_
- 10tasks_final_rlforce_rltorque_fw0.1_tw0.1_s1-20000_s2-60000_bs32_ngpu2_mlp_no_key_value_gate_no_logit_gate_
- 10tasks_final_rlforce_rltorque_fw0.1_tw0.1_s1-20000_s2-60000_bs64_ngpu4_mlp_no_key_value_gate_no_logit_gate_
- 10tasks_v2_final_base_s-60000_bs32_ngpu2
- 10tasks_v2_final_rlforce_w0.1_s1-20000_s2-60000_bs32_ngpu2_mlp_no_key_value_gate_no_logit_gate_
- 1.52 kB