Sync policy_update1175.state.json (1968376df32b)

ppo_v81/policy_update1175.state.json

@@ -0,0 +1,34 @@
+{
+  "update": 1176,
+  "stage": 5,
+  "ent_scale": 1.0,
+  "lr_now": 0.00011092632812499998,
+  "total_env_steps": 2415104,
+  "recent_wins": [
+    0.0,
+    0.0,
+    1.0,
+    0.0,
+    0.0,
+    1.0,
+    0.0
+  ],
+  "best_eval_win": 0.0,
+  "best_eval_score": 0.9322033898305084,
+  "v81_curriculum": true,
+  "stage_env_targets": [
+    24,
+    24,
+    24,
+    24,
+    24,
+    24,
+    12,
+    10,
+    8,
+    8,
+    8
+  ],
+  "envs_per_shard": 24,
+  "requested_env_target": null
+}