Run 11. Outer Step 290. Inner Step 9.

config.json

@@ -3,7 +3,7 @@
   "all_reduce_scores": {
     "0": "NOT_ALIVE",
     "1": "NOT_ALIVE",
-    "10": "SUCCESS",
+    "10": "NOT_ALIVE",
     "100": "NOT_ALIVE",
     "101": "NON_PARTICIPATING",
     "102": "NOT_ALIVE",
@@ -12,7 +12,7 @@
     "105": "NON_PARTICIPATING",
     "106": "NOT_ALIVE",
     "107": "SUCCESS",
-    "108": "NON_PARTICIPATING",
+    "108": "SUCCESS",
     "109": "SUCCESS",
     "11": "NOT_ALIVE",
     "110": "NON_PARTICIPATING",
@@ -27,7 +27,7 @@
     "119": "SUCCESS",
     "12": "NOT_ALIVE",
     "120": "NOT_ALIVE",
-    "121": "SUCCESS",
+    "121": "FAIL",
     "122": "NOT_ALIVE",
     "123": "SUCCESS",
     "124": "NOT_ALIVE",
@@ -66,7 +66,7 @@
     "154": "NOT_ALIVE",
     "155": "NOT_ALIVE",
     "156": "NOT_ALIVE",
-    "157": "SUCCESS",
+    "157": "FAIL",
     "158": "NOT_ALIVE",
     "159": "NOT_ALIVE",
     "16": "NOT_ALIVE",
@@ -79,7 +79,7 @@
     "166": "NOT_ALIVE",
     "167": "NOT_ALIVE",
     "168": "NOT_ALIVE",
-    "169": "NON_PARTICIPATING",
+    "169": "SUCCESS",
     "17": "NOT_ALIVE",
     "170": "NOT_ALIVE",
     "171": "NOT_ALIVE",
@@ -98,7 +98,7 @@
     "183": "NOT_ALIVE",
     "184": "SUCCESS",
     "185": "SUCCESS",
-    "186": "NON_PARTICIPATING",
+    "186": "SUCCESS",
     "187": "SUCCESS",
     "188": "NOT_ALIVE",
     "189": "NOT_ALIVE",
@@ -106,7 +106,7 @@
     "190": "NOT_ALIVE",
     "191": "NOT_ALIVE",
     "192": "NOT_ALIVE",
-    "193": "SUCCESS",
+    "193": "NON_PARTICIPATING",
     "194": "NOT_ALIVE",
     "195": "NON_PARTICIPATING",
     "196": "NOT_ALIVE",
@@ -130,8 +130,8 @@
     "211": "SUCCESS",
     "212": "NOT_ALIVE",
     "213": "NON_PARTICIPATING",
-    "214": "NOT_ALIVE",
-    "215": "NON_PARTICIPATING",
+    "214": "NON_PARTICIPATING",
+    "215": "SUCCESS",
     "216": "NOT_ALIVE",
     "217": "NOT_ALIVE",
     "218": "NOT_ALIVE",
@@ -149,16 +149,16 @@
     "229": "NOT_ALIVE",
     "23": "NON_PARTICIPATING",
     "230": "SUCCESS",
-    "231": "NON_PARTICIPATING",
+    "231": "SUCCESS",
     "232": "NOT_ALIVE",
-    "233": "SUCCESS",
+    "233": "FAIL",
     "234": "NON_PARTICIPATING",
     "235": "NON_PARTICIPATING",
     "236": "SUCCESS",
     "237": "SUCCESS",
     "238": "SUCCESS",
     "239": "NOT_ALIVE",
-    "24": "SUCCESS",
+    "24": "NON_PARTICIPATING",
     "240": "NOT_ALIVE",
     "241": "NON_PARTICIPATING",
     "242": "NOT_ALIVE",
@@ -171,7 +171,7 @@
     "249": "NOT_ALIVE",
     "25": "SUCCESS",
     "250": "NOT_ALIVE",
-    "251": "NOT_ALIVE",
+    "251": "NON_PARTICIPATING",
     "252": "SUCCESS",
     "253": "NOT_ALIVE",
     "254": "NOT_ALIVE",
@@ -216,10 +216,10 @@
     "6": "NOT_ALIVE",
     "60": "NON_PARTICIPATING",
     "61": "NON_PARTICIPATING",
-    "62": "NOT_ALIVE",
+    "62": "NON_PARTICIPATING",
     "63": "NOT_ALIVE",
     "64": "NOT_ALIVE",
-    "65": "SUCCESS",
+    "65": "NON_PARTICIPATING",
     "66": "NOT_ALIVE",
     "67": "NOT_ALIVE",
     "68": "SUCCESS",
@@ -230,7 +230,7 @@
     "72": "NOT_ALIVE",
     "73": "SUCCESS",
     "74": "NOT_ALIVE",
-    "75": "SUCCESS",
+    "75": "FAIL",
     "76": "NOT_ALIVE",
     "77": "NOT_ALIVE",
     "78": "NOT_ALIVE",
@@ -243,8 +243,8 @@
     "84": "NON_PARTICIPATING",
     "85": "NOT_ALIVE",
     "86": "NOT_ALIVE",
-    "87": "SUCCESS",
-    "88": "NON_PARTICIPATING",
+    "87": "NON_PARTICIPATING",
+    "88": "NOT_ALIVE",
     "89": "NON_PARTICIPATING",
     "9": "NOT_ALIVE",
     "90": "NOT_ALIVE",
@@ -256,7 +256,7 @@
     "96": "NOT_ALIVE",
     "97": "NON_PARTICIPATING",
     "98": "SUCCESS",
-    "99": "SUCCESS"
+    "99": "NON_PARTICIPATING"
   },
   "architectures": [
     "LlamaForCausalLM"
@@ -264,8 +264,8 @@
   "attention_bias": false,
   "attention_dropout": 0.0,
   "block_list": [
-    6695076,
-    6695104
+    6695210,
+    6695239
   ],
   "bos_token_id": 1,
   "dtype": "float32",
@@ -274,9 +274,9 @@
   "hidden_act": "silu",
   "hidden_size": 2048,
   "initializer_range": 0.02,
-  "inner_step": 44,
+  "inner_step": 8,
   "intermediate_size": 5632,
-  "last_allreduce_block": 6694764,
+  "last_allreduce_block": 6695164,
   "max_position_embeddings": 2048,
   "mlp_bias": false,
   "model_type": "llama",

gradients.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55b9d42c67541748ddab72cebc56cf569d8e9a9584d0f4dda85154cef9304fee
+oid sha256:50ffe35d21c2fe1e5c3a4075aea7521e659cbad3def49f773e4dd240bb112e81
 size 88389626

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56e92f2891973dde217a0191a1cdfe02ae77e10dd61b5de32bb332363363b009
+oid sha256:068913a8a817e51865b4f671189a594268bd58f7d0c0467645155c75f87e3d18
 size 4924926906

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2762212f37bf381407c58158e77d9ba2bac4033e98a73d286f77851e02e97549
+oid sha256:0254627e047d7af50878f2cb4eff79097e5298ba97484a331dd82c92efe5cc9e
 size 4400216536