Upload GPTOptim

config.json

@@ -1,25 +1,271 @@
 {
-  "_name_or_path": "distributed/optimized-gpt2-2b-vtestnet-v1",
+  "_name_or_path": "distributed/optimized-gpt2-2b",
   "activation_function": "gelu_new",
   "all_reduce_scores": {
-    "0": "SUCCESS",
-    "1": "SUCCESS",
-    "2": "SUCCESS",
-    "3": "NON_PARTICIPATING",
-    "4": "NON_PARTICIPATING",
-    "5": "NON_PARTICIPATING",
+    "0": "NON_PARTICIPATING",
+    "1": "NON_PARTICIPATING",
+    "10": "NON_PARTICIPATING",
+    "100": "NON_PARTICIPATING",
+    "101": "SUCCESS",
+    "102": "NON_PARTICIPATING",
+    "103": "SUCCESS",
+    "104": "NON_PARTICIPATING",
+    "105": "NON_PARTICIPATING",
+    "106": "NON_PARTICIPATING",
+    "107": "NON_PARTICIPATING",
+    "108": "SUCCESS",
+    "109": "SUCCESS",
+    "11": "NON_PARTICIPATING",
+    "110": "SUCCESS",
+    "111": "NON_PARTICIPATING",
+    "112": "SUCCESS",
+    "113": "NON_PARTICIPATING",
+    "114": "SUCCESS",
+    "115": "SUCCESS",
+    "116": "NON_PARTICIPATING",
+    "117": "SUCCESS",
+    "118": "SUCCESS",
+    "119": "SUCCESS",
+    "12": "NON_PARTICIPATING",
+    "120": "SUCCESS",
+    "121": "SUCCESS",
+    "122": "NON_PARTICIPATING",
+    "123": "NON_PARTICIPATING",
+    "124": "NON_PARTICIPATING",
+    "125": "NON_PARTICIPATING",
+    "126": "NON_PARTICIPATING",
+    "127": "NON_PARTICIPATING",
+    "128": "NON_PARTICIPATING",
+    "129": "NON_PARTICIPATING",
+    "13": "NON_PARTICIPATING",
+    "130": "NON_PARTICIPATING",
+    "131": "NON_PARTICIPATING",
+    "132": "SUCCESS",
+    "133": "NON_PARTICIPATING",
+    "134": "NON_PARTICIPATING",
+    "135": "SUCCESS",
+    "136": "NON_PARTICIPATING",
+    "137": "NON_PARTICIPATING",
+    "138": "NON_PARTICIPATING",
+    "139": "NON_PARTICIPATING",
+    "14": "NON_PARTICIPATING",
+    "140": "SUCCESS",
+    "141": "SUCCESS",
+    "142": "NON_PARTICIPATING",
+    "143": "SUCCESS",
+    "144": "NON_PARTICIPATING",
+    "145": "NON_PARTICIPATING",
+    "146": "NON_PARTICIPATING",
+    "147": "SUCCESS",
+    "148": "SUCCESS",
+    "149": "SUCCESS",
+    "15": "NON_PARTICIPATING",
+    "150": "SUCCESS",
+    "151": "SUCCESS",
+    "152": "NON_PARTICIPATING",
+    "153": "SUCCESS",
+    "154": "NON_PARTICIPATING",
+    "155": "NON_PARTICIPATING",
+    "156": "SUCCESS",
+    "157": "SUCCESS",
+    "158": "NON_PARTICIPATING",
+    "159": "NON_PARTICIPATING",
+    "16": "NON_PARTICIPATING",
+    "160": "NON_PARTICIPATING",
+    "161": "NON_PARTICIPATING",
+    "162": "NON_PARTICIPATING",
+    "163": "SUCCESS",
+    "164": "NON_PARTICIPATING",
+    "165": "SUCCESS",
+    "166": "NON_PARTICIPATING",
+    "167": "NON_PARTICIPATING",
+    "168": "NON_PARTICIPATING",
+    "169": "NON_PARTICIPATING",
+    "17": "NON_PARTICIPATING",
+    "170": "NON_PARTICIPATING",
+    "171": "NON_PARTICIPATING",
+    "172": "NON_PARTICIPATING",
+    "173": "NON_PARTICIPATING",
+    "174": "SUCCESS",
+    "175": "SUCCESS",
+    "176": "NON_PARTICIPATING",
+    "177": "NON_PARTICIPATING",
+    "178": "NON_PARTICIPATING",
+    "179": "NON_PARTICIPATING",
+    "18": "NON_PARTICIPATING",
+    "180": "NON_PARTICIPATING",
+    "181": "NON_PARTICIPATING",
+    "182": "NON_PARTICIPATING",
+    "183": "SUCCESS",
+    "184": "SUCCESS",
+    "185": "NON_PARTICIPATING",
+    "186": "NON_PARTICIPATING",
+    "187": "NON_PARTICIPATING",
+    "188": "NON_PARTICIPATING",
+    "189": "NON_PARTICIPATING",
+    "19": "SUCCESS",
+    "190": "NON_PARTICIPATING",
+    "191": "SUCCESS",
+    "192": "SUCCESS",
+    "193": "SUCCESS",
+    "194": "SUCCESS",
+    "195": "SUCCESS",
+    "196": "NON_PARTICIPATING",
+    "197": "NON_PARTICIPATING",
+    "198": "NON_PARTICIPATING",
+    "199": "NON_PARTICIPATING",
+    "2": "NON_PARTICIPATING",
+    "20": "NON_PARTICIPATING",
+    "200": "NON_PARTICIPATING",
+    "201": "SUCCESS",
+    "202": "NON_PARTICIPATING",
+    "203": "SUCCESS",
+    "204": "NON_PARTICIPATING",
+    "205": "SUCCESS",
+    "206": "SUCCESS",
+    "207": "NON_PARTICIPATING",
+    "208": "NON_PARTICIPATING",
+    "209": "SUCCESS",
+    "21": "NON_PARTICIPATING",
+    "210": "SUCCESS",
+    "211": "NON_PARTICIPATING",
+    "212": "NON_PARTICIPATING",
+    "213": "NON_PARTICIPATING",
+    "214": "NON_PARTICIPATING",
+    "215": "NON_PARTICIPATING",
+    "216": "NON_PARTICIPATING",
+    "217": "SUCCESS",
+    "218": "SUCCESS",
+    "219": "SUCCESS",
+    "22": "SUCCESS",
+    "220": "NON_PARTICIPATING",
+    "221": "NON_PARTICIPATING",
+    "222": "NON_PARTICIPATING",
+    "223": "NON_PARTICIPATING",
+    "224": "NON_PARTICIPATING",
+    "225": "NON_PARTICIPATING",
+    "226": "NON_PARTICIPATING",
+    "227": "NON_PARTICIPATING",
+    "228": "NON_PARTICIPATING",
+    "229": "NON_PARTICIPATING",
+    "23": "NON_PARTICIPATING",
+    "230": "SUCCESS",
+    "231": "SUCCESS",
+    "232": "NON_PARTICIPATING",
+    "233": "SUCCESS",
+    "234": "NON_PARTICIPATING",
+    "235": "SUCCESS",
+    "236": "NON_PARTICIPATING",
+    "237": "NON_PARTICIPATING",
+    "238": "SUCCESS",
+    "239": "SUCCESS",
+    "24": "SUCCESS",
+    "240": "SUCCESS",
+    "241": "SUCCESS",
+    "242": "SUCCESS",
+    "243": "NON_PARTICIPATING",
+    "244": "NON_PARTICIPATING",
+    "245": "SUCCESS",
+    "246": "NON_PARTICIPATING",
+    "247": "NON_PARTICIPATING",
+    "248": "SUCCESS",
+    "249": "NON_PARTICIPATING",
+    "25": "NON_PARTICIPATING",
+    "250": "SUCCESS",
+    "251": "NON_PARTICIPATING",
+    "252": "NON_PARTICIPATING",
+    "253": "SUCCESS",
+    "254": "NON_PARTICIPATING",
+    "255": "NON_PARTICIPATING",
+    "26": "NON_PARTICIPATING",
+    "27": "NON_PARTICIPATING",
+    "28": "SUCCESS",
+    "29": "SUCCESS",
+    "3": "SUCCESS",
+    "30": "NON_PARTICIPATING",
+    "31": "NON_PARTICIPATING",
+    "32": "NON_PARTICIPATING",
+    "33": "NON_PARTICIPATING",
+    "34": "SUCCESS",
+    "35": "SUCCESS",
+    "36": "NON_PARTICIPATING",
+    "37": "NON_PARTICIPATING",
+    "38": "NON_PARTICIPATING",
+    "39": "SUCCESS",
+    "4": "SUCCESS",
+    "40": "NON_PARTICIPATING",
+    "41": "NON_PARTICIPATING",
+    "42": "NON_PARTICIPATING",
+    "43": "SUCCESS",
+    "44": "NON_PARTICIPATING",
+    "45": "SUCCESS",
+    "46": "NON_PARTICIPATING",
+    "47": "SUCCESS",
+    "48": "NON_PARTICIPATING",
+    "49": "SUCCESS",
+    "5": "SUCCESS",
+    "50": "NON_PARTICIPATING",
+    "51": "NON_PARTICIPATING",
+    "52": "SUCCESS",
+    "53": "SUCCESS",
+    "54": "SUCCESS",
+    "55": "SUCCESS",
+    "56": "NON_PARTICIPATING",
+    "57": "SUCCESS",
+    "58": "NON_PARTICIPATING",
+    "59": "SUCCESS",
     "6": "NON_PARTICIPATING",
-    "7": "NON_PARTICIPATING",
-    "8": "NON_PARTICIPATING",
-    "9": "NON_PARTICIPATING"
+    "60": "NON_PARTICIPATING",
+    "61": "SUCCESS",
+    "62": "SUCCESS",
+    "63": "NON_PARTICIPATING",
+    "64": "NON_PARTICIPATING",
+    "65": "SUCCESS",
+    "66": "SUCCESS",
+    "67": "NON_PARTICIPATING",
+    "68": "NON_PARTICIPATING",
+    "69": "SUCCESS",
+    "7": "SUCCESS",
+    "70": "NON_PARTICIPATING",
+    "71": "SUCCESS",
+    "72": "SUCCESS",
+    "73": "SUCCESS",
+    "74": "NON_PARTICIPATING",
+    "75": "NON_PARTICIPATING",
+    "76": "NON_PARTICIPATING",
+    "77": "SUCCESS",
+    "78": "SUCCESS",
+    "79": "NON_PARTICIPATING",
+    "8": "SUCCESS",
+    "80": "NON_PARTICIPATING",
+    "81": "NON_PARTICIPATING",
+    "82": "NON_PARTICIPATING",
+    "83": "NON_PARTICIPATING",
+    "84": "NON_PARTICIPATING",
+    "85": "NON_PARTICIPATING",
+    "86": "NON_PARTICIPATING",
+    "87": "NON_PARTICIPATING",
+    "88": "SUCCESS",
+    "89": "NON_PARTICIPATING",
+    "9": "SUCCESS",
+    "90": "NON_PARTICIPATING",
+    "91": "SUCCESS",
+    "92": "NON_PARTICIPATING",
+    "93": "NON_PARTICIPATING",
+    "94": "SUCCESS",
+    "95": "NON_PARTICIPATING",
+    "96": "SUCCESS",
+    "97": "NON_PARTICIPATING",
+    "98": "NON_PARTICIPATING",
+    "99": "NON_PARTICIPATING"
   },
   "architectures": [
     "GPTOptim"
   ],
   "attn_pdrop": 0.1,
   "auto_map": {
-    "AutoConfig": "distributed/optimized-gpt2-2b-vtestnet-v1--configuration_gpt_optimized.GPTOptimConfig",
-    "AutoModelForCausalLM": "distributed/optimized-gpt2-2b-vtestnet-v1--modeling_gpt_optimized.GPTOptim"
+    "AutoConfig": "distributed/optimized-gpt2-2b--configuration_gpt_optimized.GPTOptimConfig",
+    "AutoModelForCausalLM": "distributed/optimized-gpt2-2b--modeling_gpt_optimized.GPTOptim"
   },
   "block_size": 1024,
   "bos_token_id": 50256,

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4794f690bbd12b3b33dcbcc54c0be469e419608b3fc399fd4ef5a148337864fe
+oid sha256:e04b96be504264a4e74e4ba953b6ee486ddae4808d1c8ed97bfeda269ad0991d
 size 4975530824

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22d32163ef97259d081c6a32178748c7663d12165742325c09323bd854978bab
+oid sha256:dc3e0dbd9ee3b65997a8b1f7e6c4a45c8da7cd2f44e5ee8b45e7c8bce1242828
 size 3075942944