Navigation on FrozenLake 4x4
0.95RewardASK (32B)
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| ASK (32B)Model Size=32B, trained_on=8x82026.04 | 0.95 | 7.7 | 100 | 46.17 | |
| ASK (72B)Model Size=72B, trained_on=8x82026.04 | 0.95 | 7.99 | 100 | 64.22 | |
| ASK (1.5B)Model Size=1.5B, trained_on=8x82026.04 | 0.37 | 6.92 | 100 | 42.98 | |
| ASK (14B)Model Size=14B, trained_on=8x82026.04 | 0.07 | 56.59 | 100 | 52.95 | |
| PPOtrained_on=8x82026.04 | 0 | 46.98 | — | — | |
| ASK (0.5B)Model Size=0.5B, trained_on=8x82026.04 | 0 | 41.12 | 100 | 61.71 | |
| ASK (3B)Model Size=3B, trained_on=8x82026.04 | 0 | 44.58 | 100 | 64.52 | |
| ASK (7B)Model Size=7B, trained_on=8x82026.04 | 0 | 61 | 100 | 95.57 |