first test run on train_q_rag_log.py

log_test_50.txt

@@ -0,0 +1,20 @@
+100/80000, reward=0.378, eval_reward=0.567, qf_loss=22.837, step=1313
+200/80000, reward=0.446, eval_reward=0.567, qf_loss=23.204, step=2613
+300/80000, reward=0.465, eval_reward=0.567, qf_loss=24.270, step=3913
+400/80000, reward=0.468, eval_reward=0.567, qf_loss=16.535, step=5213
+500/80000, reward=0.462, eval_reward=0.497, qf_loss=10.573, step=6513
+600/80000, reward=0.438, eval_reward=0.497, qf_loss=7.781, step=7813
+700/80000, reward=0.430, eval_reward=0.497, qf_loss=5.375, step=9113
+800/80000, reward=0.398, eval_reward=0.497, qf_loss=2.749, step=10413
+900/80000, reward=0.214, eval_reward=0.357, qf_loss=0.695, step=11713
+1000/80000, reward=0.193, eval_reward=0.357, qf_loss=0.468, step=13013
+1100/80000, reward=0.188, eval_reward=0.357, qf_loss=0.201, step=14313
+1200/80000, reward=0.178, eval_reward=0.357, qf_loss=0.181, step=15613
+1300/80000, reward=0.140, eval_reward=0.190, qf_loss=0.077, step=16913
+1400/80000, reward=0.147, eval_reward=0.190, qf_loss=0.107, step=18213
+1500/80000, reward=0.155, eval_reward=0.190, qf_loss=0.096, step=19513
+1600/80000, reward=0.165, eval_reward=0.190, qf_loss=0.093, step=20813
+1700/80000, reward=0.178, eval_reward=0.273, qf_loss=0.145, step=22113
+1800/80000, reward=0.222, eval_reward=0.273, qf_loss=0.092, step=23413
+1900/80000, reward=0.223, eval_reward=0.273, qf_loss=0.080, step=24713
+2000/80000, reward=0.245, eval_reward=0.273, qf_loss=0.081, step=26013