slot 0 carry torch.bfloat16 (2, 4, 8) a2edf5bad1fd6ba677e9bb395ed75a54141061ca369270f7db7503368bceae94
slot 0 draws torch.int64 (4,) 57e4d1a3e55577e929c7d374d3aeb7a4a287507c0748bc3c35d8ac881347a369
slot 0 observations torch.bfloat16 (2, 4, 843) 2a20fde58041ec1568ef6d9132d7ac4120fed337dcdc6b30dca88242b6d99de7
slot 0 actions torch.float32 (2, 4) d45b8aba684a9fd2fb3992aa1255f97c938bccdce78022c7db79d141d287fb2f
slot 0 logprobs torch.bfloat16 (2, 4) 9d2e1e848544a12b893275b719bf608036a20bad7f68e8faba2e9f82525ad641
slot 0 values torch.bfloat16 (2, 4) 4dab4e6ecfd495ee8a5fc58d2b88067c57e8b9145cce0cc0137706f03e02f691
slot 0 rewards torch.bfloat16 (2, 4) 7ab8f6c26e4f9862c95a18c8e5c50403eeb64d8869fbbf9a7a6397d9a63b7b0e
slot 0 terminals torch.bfloat16 (2, 4) 374708fff7719dd5979ec875d56cd2286f6d3cf7ec317a3b25632aab28ec37bb
slot 0 action_mask torch.bfloat16 (2, 4, 43) 84bd54223e5ad0703711d31c3cdc824c1ce87e7e757ae978ff70604c72f35b36
slot 0 initial_states torch.bfloat16 (2, 4, 8) 38723a2e5e8a17aa7950dc008209944e898f69a7bd10a23c839d341e935fd5ca
slot 1 carry torch.bfloat16 (2, 4, 8) 25f570d46cd3b540f4b80efc9c8434db4cac31231a53a06e44ed8d0c3503cfbc
slot 1 draws torch.int64 (4,) 38df89b71e6aa2b361da02c1960692788be8f78764a50866b24f7c7710cb0dbe
slot 1 observations torch.bfloat16 (2, 4, 843) de527badb0ca7c77ad5d9d2ff056007343acdcd874f3d30ec608cf863c888fac
slot 1 actions torch.float32 (2, 4) 0d1c3defea040cd114c81a9bd5f6dd59623599943da3fb10177766a155157a34
slot 1 logprobs torch.bfloat16 (2, 4) 80ce71844fe0bc52d2a0c6382ae5bff05dea5e15f22c4939d3b0f7fa56c6e46d
slot 1 values torch.bfloat16 (2, 4) d559b499d5c4e7d2c49344cca78fbe653cc05974da79b2ea9376a2a18d8f80fa
slot 1 rewards torch.bfloat16 (2, 4) 374708fff7719dd5979ec875d56cd2286f6d3cf7ec317a3b25632aab28ec37bb
slot 1 terminals torch.bfloat16 (2, 4) 374708fff7719dd5979ec875d56cd2286f6d3cf7ec317a3b25632aab28ec37bb
slot 1 action_mask torch.bfloat16 (2, 4, 43) 540218d3adfb001ce89bd2580819a7956ea810cfeabb4df62632fa0e0bf716d9
slot 1 initial_states torch.bfloat16 (2, 4, 8) a2edf5bad1fd6ba677e9bb395ed75a54141061ca369270f7db7503368bceae94
states |u1 (320992,) 61b12ae06ab750d21acfd3c4bc31fe3ec577956e9610d99446156cd2ea50b82c
rngs <u4 (4,) 4e2e9c953a0259e4003f37eb57e44c8650fd297a2aac9671b270e9b20cb2740f
stats |u1 (1424,) 4a14730ac355d207cc70382dfa1c6386091c36d9d5666991796298fe4a05d3fd
observations torch.float32 (4, 843) 15e8e9f0e0d43ef3b216490884a6c70b705a9ea8bdb522bddc206ee93122294f
action_mask torch.uint8 (4, 43) a115ff43dd0cfd27ec64285cd5ea17a3c194baaca0ce93d366dd7d06a31d38d3
rewards torch.float32 (4,) 6855960ee428276148fc8745625c44bde7ae835a9e2fa181cc917af9bcad0d6a
terminals torch.float32 (4,) 374708fff7719dd5979ec875d56cd2286f6d3cf7ec317a3b25632aab28ec37bb
actions torch.float32 (4, 1) 5944e6ecb837cab7d36db247a1f06d158719bb8ed511c1a353d2e45dc72089c4
