Run 4. Outer Step 50. Inner Step 2.

config.json

@@ -3,8 +3,8 @@
   "activation_function": "gelu_new",
   "all_reduce_scores": {
     "0": "NON_PARTICIPATING",
-    "1": "NON_PARTICIPATING",
-    "10": "SUCCESS",
+    "1": "SUCCESS",
+    "10": "NON_PARTICIPATING",
     "100": "NON_PARTICIPATING",
     "101": "NON_PARTICIPATING",
     "102": "NON_PARTICIPATING",
@@ -12,65 +12,65 @@
     "104": "NON_PARTICIPATING",
     "105": "NON_PARTICIPATING",
     "106": "NON_PARTICIPATING",
-    "107": "SUCCESS",
+    "107": "NON_PARTICIPATING",
     "108": "NON_PARTICIPATING",
     "109": "NON_PARTICIPATING",
     "11": "SUCCESS",
-    "110": "SUCCESS",
+    "110": "NON_PARTICIPATING",
     "111": "NON_PARTICIPATING",
     "112": "NON_PARTICIPATING",
     "113": "NON_PARTICIPATING",
     "114": "NON_PARTICIPATING",
     "115": "NON_PARTICIPATING",
     "116": "NON_PARTICIPATING",
-    "117": "SUCCESS",
+    "117": "NON_PARTICIPATING",
     "118": "SUCCESS",
     "119": "NON_PARTICIPATING",
-    "12": "NON_PARTICIPATING",
-    "120": "SUCCESS",
+    "12": "SUCCESS",
+    "120": "NON_PARTICIPATING",
     "121": "NON_PARTICIPATING",
-    "122": "SUCCESS",
-    "123": "SUCCESS",
+    "122": "NON_PARTICIPATING",
+    "123": "NON_PARTICIPATING",
     "124": "NON_PARTICIPATING",
-    "125": "SUCCESS",
+    "125": "NON_PARTICIPATING",
     "126": "NON_PARTICIPATING",
     "127": "NON_PARTICIPATING",
     "128": "NON_PARTICIPATING",
     "129": "NON_PARTICIPATING",
-    "13": "NON_PARTICIPATING",
+    "13": "SUCCESS",
     "130": "NON_PARTICIPATING",
     "131": "NON_PARTICIPATING",
-    "132": "SUCCESS",
+    "132": "NON_PARTICIPATING",
     "133": "NON_PARTICIPATING",
-    "134": "SUCCESS",
-    "135": "SUCCESS",
-    "136": "NON_PARTICIPATING",
+    "134": "NON_PARTICIPATING",
+    "135": "NON_PARTICIPATING",
+    "136": "SUCCESS",
     "137": "NON_PARTICIPATING",
-    "138": "NON_PARTICIPATING",
+    "138": "SUCCESS",
     "139": "NON_PARTICIPATING",
-    "14": "NON_PARTICIPATING",
+    "14": "SUCCESS",
     "140": "NON_PARTICIPATING",
     "141": "NON_PARTICIPATING",
-    "142": "NON_PARTICIPATING",
+    "142": "SUCCESS",
     "143": "NON_PARTICIPATING",
     "144": "NON_PARTICIPATING",
     "145": "NON_PARTICIPATING",
     "146": "NON_PARTICIPATING",
     "147": "NON_PARTICIPATING",
-    "148": "SUCCESS",
-    "149": "SUCCESS",
-    "15": "SUCCESS",
-    "150": "SUCCESS",
+    "148": "NON_PARTICIPATING",
+    "149": "NON_PARTICIPATING",
+    "15": "NON_PARTICIPATING",
+    "150": "NON_PARTICIPATING",
     "151": "NON_PARTICIPATING",
     "152": "NON_PARTICIPATING",
-    "153": "SUCCESS",
-    "154": "SUCCESS",
+    "153": "NON_PARTICIPATING",
+    "154": "NON_PARTICIPATING",
     "155": "NON_PARTICIPATING",
-    "156": "SUCCESS",
-    "157": "NON_PARTICIPATING",
+    "156": "NON_PARTICIPATING",
+    "157": "SUCCESS",
     "158": "NON_PARTICIPATING",
-    "159": "SUCCESS",
-    "16": "NON_PARTICIPATING",
+    "159": "NON_PARTICIPATING",
+    "16": "SUCCESS",
     "160": "SUCCESS",
     "161": "NON_PARTICIPATING",
     "162": "NON_PARTICIPATING",
@@ -78,23 +78,23 @@
     "164": "NON_PARTICIPATING",
     "165": "NON_PARTICIPATING",
     "166": "NON_PARTICIPATING",
-    "167": "SUCCESS",
+    "167": "NON_PARTICIPATING",
     "168": "NON_PARTICIPATING",
     "169": "NON_PARTICIPATING",
-    "17": "NON_PARTICIPATING",
+    "17": "SUCCESS",
     "170": "NON_PARTICIPATING",
     "171": "NON_PARTICIPATING",
-    "172": "SUCCESS",
+    "172": "NON_PARTICIPATING",
     "173": "NON_PARTICIPATING",
     "174": "NON_PARTICIPATING",
     "175": "SUCCESS",
     "176": "NON_PARTICIPATING",
     "177": "SUCCESS",
-    "178": "SUCCESS",
+    "178": "NON_PARTICIPATING",
     "179": "NON_PARTICIPATING",
-    "18": "NON_PARTICIPATING",
+    "18": "SUCCESS",
     "180": "NON_PARTICIPATING",
-    "181": "NON_PARTICIPATING",
+    "181": "SUCCESS",
     "182": "NON_PARTICIPATING",
     "183": "NON_PARTICIPATING",
     "184": "NON_PARTICIPATING",
@@ -103,160 +103,160 @@
     "187": "NON_PARTICIPATING",
     "188": "NON_PARTICIPATING",
     "189": "NON_PARTICIPATING",
-    "19": "NON_PARTICIPATING",
+    "19": "SUCCESS",
     "190": "NON_PARTICIPATING",
     "191": "NON_PARTICIPATING",
-    "192": "SUCCESS",
+    "192": "NON_PARTICIPATING",
     "193": "NON_PARTICIPATING",
-    "194": "SUCCESS",
-    "195": "SUCCESS",
+    "194": "NON_PARTICIPATING",
+    "195": "NON_PARTICIPATING",
     "196": "NON_PARTICIPATING",
     "197": "NON_PARTICIPATING",
     "198": "NON_PARTICIPATING",
     "199": "NON_PARTICIPATING",
-    "2": "NON_PARTICIPATING",
-    "20": "NON_PARTICIPATING",
+    "2": "SUCCESS",
+    "20": "SUCCESS",
     "200": "NON_PARTICIPATING",
-    "201": "NON_PARTICIPATING",
-    "202": "SUCCESS",
+    "201": "SUCCESS",
+    "202": "NON_PARTICIPATING",
     "203": "SUCCESS",
     "204": "NON_PARTICIPATING",
     "205": "NON_PARTICIPATING",
-    "206": "SUCCESS",
+    "206": "NON_PARTICIPATING",
     "207": "NON_PARTICIPATING",
     "208": "NON_PARTICIPATING",
     "209": "NON_PARTICIPATING",
-    "21": "NON_PARTICIPATING",
+    "21": "SUCCESS",
     "210": "NON_PARTICIPATING",
     "211": "NON_PARTICIPATING",
     "212": "NON_PARTICIPATING",
     "213": "NON_PARTICIPATING",
-    "214": "SUCCESS",
+    "214": "NON_PARTICIPATING",
     "215": "SUCCESS",
-    "216": "SUCCESS",
+    "216": "NON_PARTICIPATING",
     "217": "NON_PARTICIPATING",
-    "218": "SUCCESS",
+    "218": "NON_PARTICIPATING",
     "219": "NON_PARTICIPATING",
     "22": "SUCCESS",
-    "220": "NON_PARTICIPATING",
+    "220": "SUCCESS",
     "221": "NON_PARTICIPATING",
-    "222": "SUCCESS",
+    "222": "NON_PARTICIPATING",
     "223": "NON_PARTICIPATING",
     "224": "NON_PARTICIPATING",
     "225": "NON_PARTICIPATING",
-    "226": "SUCCESS",
+    "226": "NON_PARTICIPATING",
     "227": "NON_PARTICIPATING",
     "228": "NON_PARTICIPATING",
-    "229": "NON_PARTICIPATING",
+    "229": "SUCCESS",
     "23": "SUCCESS",
     "230": "NON_PARTICIPATING",
     "231": "NON_PARTICIPATING",
     "232": "NON_PARTICIPATING",
-    "233": "NON_PARTICIPATING",
+    "233": "SUCCESS",
     "234": "NON_PARTICIPATING",
     "235": "NON_PARTICIPATING",
     "236": "NON_PARTICIPATING",
     "237": "NON_PARTICIPATING",
     "238": "NON_PARTICIPATING",
     "239": "NON_PARTICIPATING",
-    "24": "NON_PARTICIPATING",
+    "24": "SUCCESS",
     "240": "NON_PARTICIPATING",
-    "241": "SUCCESS",
-    "242": "NON_PARTICIPATING",
+    "241": "NON_PARTICIPATING",
+    "242": "SUCCESS",
     "243": "NON_PARTICIPATING",
     "244": "NON_PARTICIPATING",
     "245": "NON_PARTICIPATING",
     "246": "NON_PARTICIPATING",
     "247": "SUCCESS",
     "248": "NON_PARTICIPATING",
-    "249": "SUCCESS",
+    "249": "NON_PARTICIPATING",
     "25": "SUCCESS",
     "250": "NON_PARTICIPATING",
     "251": "NON_PARTICIPATING",
     "252": "NON_PARTICIPATING",
     "253": "NON_PARTICIPATING",
     "254": "NON_PARTICIPATING",
-    "255": "SUCCESS",
+    "255": "NON_PARTICIPATING",
     "26": "SUCCESS",
     "27": "SUCCESS",
-    "28": "NON_PARTICIPATING",
+    "28": "SUCCESS",
     "29": "SUCCESS",
-    "3": "NON_PARTICIPATING",
-    "30": "NON_PARTICIPATING",
+    "3": "SUCCESS",
+    "30": "SUCCESS",
     "31": "NON_PARTICIPATING",
     "32": "SUCCESS",
     "33": "NON_PARTICIPATING",
     "34": "SUCCESS",
     "35": "SUCCESS",
-    "36": "NON_PARTICIPATING",
-    "37": "NON_PARTICIPATING",
+    "36": "SUCCESS",
+    "37": "SUCCESS",
     "38": "NON_PARTICIPATING",
-    "39": "NON_PARTICIPATING",
+    "39": "SUCCESS",
     "4": "SUCCESS",
     "40": "SUCCESS",
-    "41": "NON_PARTICIPATING",
+    "41": "SUCCESS",
     "42": "NON_PARTICIPATING",
-    "43": "NON_PARTICIPATING",
-    "44": "NON_PARTICIPATING",
-    "45": "NON_PARTICIPATING",
+    "43": "SUCCESS",
+    "44": "SUCCESS",
+    "45": "SUCCESS",
     "46": "NON_PARTICIPATING",
-    "47": "NON_PARTICIPATING",
+    "47": "SUCCESS",
     "48": "NON_PARTICIPATING",
     "49": "SUCCESS",
-    "5": "NON_PARTICIPATING",
-    "50": "NON_PARTICIPATING",
+    "5": "SUCCESS",
+    "50": "SUCCESS",
     "51": "SUCCESS",
-    "52": "NON_PARTICIPATING",
-    "53": "NON_PARTICIPATING",
-    "54": "NON_PARTICIPATING",
+    "52": "SUCCESS",
+    "53": "SUCCESS",
+    "54": "SUCCESS",
     "55": "SUCCESS",
-    "56": "NON_PARTICIPATING",
-    "57": "NON_PARTICIPATING",
-    "58": "SUCCESS",
-    "59": "NON_PARTICIPATING",
-    "6": "SUCCESS",
-    "60": "NON_PARTICIPATING",
+    "56": "SUCCESS",
+    "57": "SUCCESS",
+    "58": "NON_PARTICIPATING",
+    "59": "SUCCESS",
+    "6": "NON_PARTICIPATING",
+    "60": "SUCCESS",
     "61": "NON_PARTICIPATING",
-    "62": "NON_PARTICIPATING",
+    "62": "SUCCESS",
     "63": "SUCCESS",
     "64": "NON_PARTICIPATING",
-    "65": "NON_PARTICIPATING",
-    "66": "NON_PARTICIPATING",
+    "65": "SUCCESS",
+    "66": "SUCCESS",
     "67": "SUCCESS",
-    "68": "NON_PARTICIPATING",
-    "69": "NON_PARTICIPATING",
-    "7": "NON_PARTICIPATING",
+    "68": "SUCCESS",
+    "69": "SUCCESS",
+    "7": "SUCCESS",
     "70": "NON_PARTICIPATING",
     "71": "NON_PARTICIPATING",
-    "72": "NON_PARTICIPATING",
+    "72": "SUCCESS",
     "73": "NON_PARTICIPATING",
-    "74": "NON_PARTICIPATING",
+    "74": "SUCCESS",
     "75": "SUCCESS",
-    "76": "NON_PARTICIPATING",
-    "77": "NON_PARTICIPATING",
+    "76": "SUCCESS",
+    "77": "SUCCESS",
     "78": "SUCCESS",
     "79": "NON_PARTICIPATING",
-    "8": "NON_PARTICIPATING",
+    "8": "SUCCESS",
     "80": "SUCCESS",
-    "81": "NON_PARTICIPATING",
+    "81": "SUCCESS",
     "82": "SUCCESS",
     "83": "NON_PARTICIPATING",
-    "84": "NON_PARTICIPATING",
+    "84": "SUCCESS",
     "85": "NON_PARTICIPATING",
     "86": "NON_PARTICIPATING",
     "87": "SUCCESS",
     "88": "SUCCESS",
     "89": "SUCCESS",
-    "9": "NON_PARTICIPATING",
+    "9": "SUCCESS",
     "90": "SUCCESS",
-    "91": "NON_PARTICIPATING",
+    "91": "SUCCESS",
     "92": "SUCCESS",
     "93": "NON_PARTICIPATING",
-    "94": "NON_PARTICIPATING",
+    "94": "SUCCESS",
     "95": "SUCCESS",
-    "96": "NON_PARTICIPATING",
-    "97": "NON_PARTICIPATING",
-    "98": "NON_PARTICIPATING",
+    "96": "SUCCESS",
+    "97": "SUCCESS",
+    "98": "SUCCESS",
     "99": "SUCCESS"
   },
   "architectures": [
@@ -268,20 +268,17 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5505978,
-    5505982,
-    5505986,
-    5505990,
-    5505994
+    5625110,
+    5625140
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 25,
+  "inner_step": 2,
   "inner_steps": 0,
-  "last_allreduce_block": 5503964,
+  "last_allreduce_block": 5571611,
   "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:d37a5559fa7c6edfff663ccbf762ad5ea7407b4eee91e638abcf1dfeddd9a98e
+oid sha256:8e18140bd3058afbd828f6d847221752ed39a050362b60995f00f550533ed564
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:940ce2ac9b43cf198c713138802c75b769a52a4675a67111f90bac334ae5baff
+oid sha256:69ef0292c6b3d0eaed1b99239fdd22cc2a3997bbfc6bb0e57cc8af7d4e115c62
 size 4040701744