Run 4. Outer Step 77. Inner Step 0. Peers 96.

config.json

@@ -3,10 +3,10 @@
   "activation_function": "gelu_new",
   "all_reduce_scores": {
     "0": "NON_PARTICIPATING",
-    "1": "NON_PARTICIPATING",
+    "1": "SUCCESS",
     "10": "NON_PARTICIPATING",
     "100": "NON_PARTICIPATING",
-    "101": "NON_PARTICIPATING",
+    "101": "SUCCESS",
     "102": "NON_PARTICIPATING",
     "103": "NON_PARTICIPATING",
     "104": "NON_PARTICIPATING",
@@ -15,8 +15,8 @@
     "107": "NON_PARTICIPATING",
     "108": "NON_PARTICIPATING",
     "109": "NON_PARTICIPATING",
-    "11": "NON_PARTICIPATING",
-    "110": "SUCCESS",
+    "11": "SUCCESS",
+    "110": "NON_PARTICIPATING",
     "111": "NON_PARTICIPATING",
     "112": "NON_PARTICIPATING",
     "113": "NON_PARTICIPATING",
@@ -24,31 +24,31 @@
     "115": "NON_PARTICIPATING",
     "116": "NON_PARTICIPATING",
     "117": "NON_PARTICIPATING",
-    "118": "SUCCESS",
+    "118": "NON_PARTICIPATING",
     "119": "NON_PARTICIPATING",
-    "12": "NON_PARTICIPATING",
+    "12": "SUCCESS",
     "120": "NON_PARTICIPATING",
-    "121": "NON_PARTICIPATING",
+    "121": "SUCCESS",
     "122": "NON_PARTICIPATING",
     "123": "NON_PARTICIPATING",
     "124": "NON_PARTICIPATING",
     "125": "NON_PARTICIPATING",
     "126": "NON_PARTICIPATING",
     "127": "NON_PARTICIPATING",
-    "128": "NON_PARTICIPATING",
-    "129": "SUCCESS",
-    "13": "NON_PARTICIPATING",
+    "128": "SUCCESS",
+    "129": "NON_PARTICIPATING",
+    "13": "SUCCESS",
     "130": "NON_PARTICIPATING",
     "131": "NON_PARTICIPATING",
     "132": "NON_PARTICIPATING",
-    "133": "NON_PARTICIPATING",
+    "133": "SUCCESS",
     "134": "NON_PARTICIPATING",
     "135": "NON_PARTICIPATING",
     "136": "NON_PARTICIPATING",
     "137": "NON_PARTICIPATING",
     "138": "NON_PARTICIPATING",
     "139": "NON_PARTICIPATING",
-    "14": "NON_PARTICIPATING",
+    "14": "SUCCESS",
     "140": "NON_PARTICIPATING",
     "141": "NON_PARTICIPATING",
     "142": "NON_PARTICIPATING",
@@ -56,43 +56,43 @@
     "144": "NON_PARTICIPATING",
     "145": "NON_PARTICIPATING",
     "146": "NON_PARTICIPATING",
-    "147": "SUCCESS",
+    "147": "NON_PARTICIPATING",
     "148": "NON_PARTICIPATING",
     "149": "NON_PARTICIPATING",
-    "15": "NON_PARTICIPATING",
+    "15": "SUCCESS",
     "150": "NON_PARTICIPATING",
     "151": "NON_PARTICIPATING",
     "152": "SUCCESS",
-    "153": "SUCCESS",
-    "154": "FAIL",
-    "155": "SUCCESS",
+    "153": "NON_PARTICIPATING",
+    "154": "NON_PARTICIPATING",
+    "155": "NON_PARTICIPATING",
     "156": "NON_PARTICIPATING",
     "157": "NON_PARTICIPATING",
     "158": "NON_PARTICIPATING",
     "159": "NON_PARTICIPATING",
-    "16": "NON_PARTICIPATING",
+    "16": "SUCCESS",
     "160": "NON_PARTICIPATING",
     "161": "NON_PARTICIPATING",
     "162": "NON_PARTICIPATING",
-    "163": "SUCCESS",
+    "163": "NON_PARTICIPATING",
     "164": "NON_PARTICIPATING",
     "165": "NON_PARTICIPATING",
     "166": "NON_PARTICIPATING",
     "167": "NON_PARTICIPATING",
     "168": "NON_PARTICIPATING",
     "169": "NON_PARTICIPATING",
-    "17": "NON_PARTICIPATING",
+    "17": "SUCCESS",
     "170": "NON_PARTICIPATING",
-    "171": "NON_PARTICIPATING",
+    "171": "SUCCESS",
     "172": "NON_PARTICIPATING",
     "173": "NON_PARTICIPATING",
     "174": "NON_PARTICIPATING",
     "175": "NON_PARTICIPATING",
     "176": "NON_PARTICIPATING",
-    "177": "SUCCESS",
+    "177": "NON_PARTICIPATING",
     "178": "NON_PARTICIPATING",
     "179": "NON_PARTICIPATING",
-    "18": "NON_PARTICIPATING",
+    "18": "SUCCESS",
     "180": "NON_PARTICIPATING",
     "181": "NON_PARTICIPATING",
     "182": "NON_PARTICIPATING",
@@ -103,7 +103,7 @@
     "187": "NON_PARTICIPATING",
     "188": "NON_PARTICIPATING",
     "189": "NON_PARTICIPATING",
-    "19": "NON_PARTICIPATING",
+    "19": "SUCCESS",
     "190": "NON_PARTICIPATING",
     "191": "NON_PARTICIPATING",
     "192": "NON_PARTICIPATING",
@@ -114,7 +114,7 @@
     "197": "NON_PARTICIPATING",
     "198": "NON_PARTICIPATING",
     "199": "NON_PARTICIPATING",
-    "2": "NON_PARTICIPATING",
+    "2": "SUCCESS",
     "20": "SUCCESS",
     "200": "SUCCESS",
     "201": "NON_PARTICIPATING",
@@ -125,30 +125,30 @@
     "206": "NON_PARTICIPATING",
     "207": "NON_PARTICIPATING",
     "208": "SUCCESS",
-    "209": "NON_PARTICIPATING",
-    "21": "NON_PARTICIPATING",
+    "209": "SUCCESS",
+    "21": "SUCCESS",
     "210": "NON_PARTICIPATING",
     "211": "NON_PARTICIPATING",
-    "212": "SUCCESS",
-    "213": "SUCCESS",
+    "212": "NON_PARTICIPATING",
+    "213": "NON_PARTICIPATING",
     "214": "NON_PARTICIPATING",
     "215": "NON_PARTICIPATING",
-    "216": "NON_PARTICIPATING",
+    "216": "SUCCESS",
     "217": "NON_PARTICIPATING",
     "218": "NON_PARTICIPATING",
     "219": "NON_PARTICIPATING",
-    "22": "NON_PARTICIPATING",
+    "22": "SUCCESS",
     "220": "NON_PARTICIPATING",
     "221": "NON_PARTICIPATING",
-    "222": "NON_PARTICIPATING",
+    "222": "SUCCESS",
     "223": "NON_PARTICIPATING",
     "224": "NON_PARTICIPATING",
     "225": "NON_PARTICIPATING",
     "226": "NON_PARTICIPATING",
-    "227": "NON_PARTICIPATING",
-    "228": "SUCCESS",
+    "227": "SUCCESS",
+    "228": "NON_PARTICIPATING",
     "229": "NON_PARTICIPATING",
-    "23": "NON_PARTICIPATING",
+    "23": "SUCCESS",
     "230": "NON_PARTICIPATING",
     "231": "NON_PARTICIPATING",
     "232": "NON_PARTICIPATING",
@@ -159,105 +159,105 @@
     "237": "NON_PARTICIPATING",
     "238": "NON_PARTICIPATING",
     "239": "NON_PARTICIPATING",
-    "24": "NON_PARTICIPATING",
+    "24": "SUCCESS",
     "240": "NON_PARTICIPATING",
     "241": "NON_PARTICIPATING",
     "242": "NON_PARTICIPATING",
-    "243": "NON_PARTICIPATING",
+    "243": "SUCCESS",
     "244": "NON_PARTICIPATING",
     "245": "NON_PARTICIPATING",
     "246": "NON_PARTICIPATING",
-    "247": "SUCCESS",
+    "247": "NON_PARTICIPATING",
     "248": "NON_PARTICIPATING",
     "249": "NON_PARTICIPATING",
-    "25": "SUCCESS",
+    "25": "NON_PARTICIPATING",
     "250": "NON_PARTICIPATING",
     "251": "NON_PARTICIPATING",
     "252": "NON_PARTICIPATING",
     "253": "NON_PARTICIPATING",
     "254": "NON_PARTICIPATING",
-    "255": "NON_PARTICIPATING",
+    "255": "SUCCESS",
     "26": "NON_PARTICIPATING",
-    "27": "NON_PARTICIPATING",
-    "28": "NON_PARTICIPATING",
-    "29": "NON_PARTICIPATING",
-    "3": "NON_PARTICIPATING",
+    "27": "SUCCESS",
+    "28": "SUCCESS",
+    "29": "SUCCESS",
+    "3": "SUCCESS",
     "30": "NON_PARTICIPATING",
-    "31": "NON_PARTICIPATING",
-    "32": "NON_PARTICIPATING",
+    "31": "SUCCESS",
+    "32": "SUCCESS",
     "33": "NON_PARTICIPATING",
     "34": "NON_PARTICIPATING",
-    "35": "NON_PARTICIPATING",
-    "36": "NON_PARTICIPATING",
-    "37": "NON_PARTICIPATING",
-    "38": "NON_PARTICIPATING",
-    "39": "NON_PARTICIPATING",
-    "4": "NON_PARTICIPATING",
-    "40": "NON_PARTICIPATING",
-    "41": "NON_PARTICIPATING",
-    "42": "NON_PARTICIPATING",
-    "43": "NON_PARTICIPATING",
-    "44": "NON_PARTICIPATING",
-    "45": "NON_PARTICIPATING",
+    "35": "SUCCESS",
+    "36": "SUCCESS",
+    "37": "SUCCESS",
+    "38": "SUCCESS",
+    "39": "SUCCESS",
+    "4": "SUCCESS",
+    "40": "SUCCESS",
+    "41": "SUCCESS",
+    "42": "SUCCESS",
+    "43": "SUCCESS",
+    "44": "SUCCESS",
+    "45": "SUCCESS",
     "46": "NON_PARTICIPATING",
-    "47": "NON_PARTICIPATING",
+    "47": "SUCCESS",
     "48": "NON_PARTICIPATING",
-    "49": "NON_PARTICIPATING",
-    "5": "NON_PARTICIPATING",
-    "50": "NON_PARTICIPATING",
-    "51": "NON_PARTICIPATING",
-    "52": "NON_PARTICIPATING",
-    "53": "NON_PARTICIPATING",
-    "54": "NON_PARTICIPATING",
-    "55": "NON_PARTICIPATING",
-    "56": "NON_PARTICIPATING",
+    "49": "SUCCESS",
+    "5": "SUCCESS",
+    "50": "SUCCESS",
+    "51": "SUCCESS",
+    "52": "SUCCESS",
+    "53": "SUCCESS",
+    "54": "SUCCESS",
+    "55": "SUCCESS",
+    "56": "SUCCESS",
     "57": "NON_PARTICIPATING",
     "58": "NON_PARTICIPATING",
     "59": "NON_PARTICIPATING",
-    "6": "NON_PARTICIPATING",
+    "6": "SUCCESS",
     "60": "NON_PARTICIPATING",
-    "61": "NON_PARTICIPATING",
-    "62": "NON_PARTICIPATING",
-    "63": "NON_PARTICIPATING",
-    "64": "SUCCESS",
-    "65": "NON_PARTICIPATING",
-    "66": "NON_PARTICIPATING",
-    "67": "NON_PARTICIPATING",
-    "68": "NON_PARTICIPATING",
-    "69": "NON_PARTICIPATING",
-    "7": "NON_PARTICIPATING",
-    "70": "NON_PARTICIPATING",
+    "61": "SUCCESS",
+    "62": "SUCCESS",
+    "63": "SUCCESS",
+    "64": "NON_PARTICIPATING",
+    "65": "SUCCESS",
+    "66": "SUCCESS",
+    "67": "SUCCESS",
+    "68": "SUCCESS",
+    "69": "SUCCESS",
+    "7": "SUCCESS",
+    "70": "SUCCESS",
     "71": "SUCCESS",
-    "72": "NON_PARTICIPATING",
-    "73": "NON_PARTICIPATING",
-    "74": "NON_PARTICIPATING",
+    "72": "SUCCESS",
+    "73": "SUCCESS",
+    "74": "SUCCESS",
     "75": "NON_PARTICIPATING",
-    "76": "NON_PARTICIPATING",
+    "76": "SUCCESS",
     "77": "SUCCESS",
-    "78": "NON_PARTICIPATING",
+    "78": "SUCCESS",
     "79": "NON_PARTICIPATING",
-    "8": "NON_PARTICIPATING",
-    "80": "NON_PARTICIPATING",
-    "81": "NON_PARTICIPATING",
-    "82": "NON_PARTICIPATING",
-    "83": "NON_PARTICIPATING",
-    "84": "NON_PARTICIPATING",
+    "8": "SUCCESS",
+    "80": "SUCCESS",
+    "81": "SUCCESS",
+    "82": "SUCCESS",
+    "83": "SUCCESS",
+    "84": "SUCCESS",
     "85": "NON_PARTICIPATING",
     "86": "NON_PARTICIPATING",
-    "87": "NON_PARTICIPATING",
-    "88": "NON_PARTICIPATING",
-    "89": "NON_PARTICIPATING",
-    "9": "NON_PARTICIPATING",
-    "90": "NON_PARTICIPATING",
-    "91": "NON_PARTICIPATING",
-    "92": "NON_PARTICIPATING",
+    "87": "SUCCESS",
+    "88": "SUCCESS",
+    "89": "SUCCESS",
+    "9": "SUCCESS",
+    "90": "SUCCESS",
+    "91": "SUCCESS",
+    "92": "SUCCESS",
     "93": "NON_PARTICIPATING",
-    "94": "NON_PARTICIPATING",
+    "94": "SUCCESS",
     "95": "NON_PARTICIPATING",
-    "96": "NON_PARTICIPATING",
-    "97": "NON_PARTICIPATING",
-    "98": "NON_PARTICIPATING",
-    "99": "NON_PARTICIPATING"
+    "96": "SUCCESS",
+    "97": "SUCCESS",
+    "98": "SUCCESS",
+    "99": "SUCCESS"
   },
   "architectures": [
     "GPTOptim"
@@ -275,7 +275,7 @@
   "initializer_range": 0.02,
   "inner_step": 0,
   "inner_steps": 0,
-  "last_allreduce_block": 5663576,
+  "last_allreduce_block": 5664450,
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt_optimized",
   "n_embd": 1280,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c922399cbc4848bcd459bdcbc3bb3a18f2168206d0d1ff14460fdd49f221f91
+oid sha256:c6d99f676b45791f1b794325a41e8c15151f5981f854077a8b6bf41bc7af85ce
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72a8c5a75caa0967e4dbb4ff67a743437dd6a62032df6090255e09c84d3138bb
+oid sha256:19135df2a5acec63caf4d2b0d33ada37675f4e09f9239997f0b89a930a8ab15e
 size 4040701744

outer_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:221a95173469ad6adb59dd0468ca54506adc6e1da8fcf3030c09536266b22fd3
+oid sha256:2a1c81e7d163226cd7b3423feab7f9e11b2c1e925b5656682ab5ce62b55ebc5e
 size 4040805354