python -fromjson  -name correctness -v 1 -a ppo -dist clifford --wandb -testfile random-test-3q -seed 42 -qbits 3 -tol 0.1 -gamma 0.99 -numeps 50 -maxsteps 20 -rewardtype 4 -noise 0.0 -batchsz 32 -terct 100 -phidden (128, 128) -vhidden (128, 128) -plr 0.001 -vlr 0.001 -activfn relu -polyak 0.005 -meanbd 1.0 -stdtol 0.1 -numworkers 1 -entropywt 0.001 -tau 0.97 -expltime 1.0 -alpha 0.3 -beta0 0.6 -betarate 0.9992 -bufsize 256 -num_envs 8 -exptdate 30-09-2024 -gateset ('h', 's', 'cnot', 'hsh') 
