Run 4. Outer Step 6. Inner Step 0.

config.json

@@ -1,16 +1,16 @@
 {
-  "_name_or_path": "wandb02/c66-h13",
+  "_name_or_path": "distributed/optimized-gpt2-1b",
   "activation_function": "gelu_new",
   "all_reduce_scores": {
     "0": "NON_PARTICIPATING",
-    "1": "NON_PARTICIPATING",
+    "1": "SUCCESS",
     "10": "SUCCESS",
     "100": "NON_PARTICIPATING",
     "101": "SUCCESS",
     "102": "NON_PARTICIPATING",
     "103": "NON_PARTICIPATING",
     "104": "NON_PARTICIPATING",
-    "105": "NON_PARTICIPATING",
+    "105": "SUCCESS",
     "106": "NON_PARTICIPATING",
     "107": "SUCCESS",
     "108": "NON_PARTICIPATING",
@@ -20,7 +20,7 @@
     "111": "NON_PARTICIPATING",
     "112": "NON_PARTICIPATING",
     "113": "NON_PARTICIPATING",
-    "114": "SUCCESS",
+    "114": "NON_PARTICIPATING",
     "115": "NON_PARTICIPATING",
     "116": "NON_PARTICIPATING",
     "117": "NON_PARTICIPATING",
@@ -58,7 +58,7 @@
     "146": "NON_PARTICIPATING",
     "147": "SUCCESS",
     "148": "SUCCESS",
-    "149": "NON_PARTICIPATING",
+    "149": "SUCCESS",
     "15": "SUCCESS",
     "150": "NON_PARTICIPATING",
     "151": "NON_PARTICIPATING",
@@ -71,7 +71,7 @@
     "158": "NON_PARTICIPATING",
     "159": "NON_PARTICIPATING",
     "16": "SUCCESS",
-    "160": "SUCCESS",
+    "160": "FAIL",
     "161": "NON_PARTICIPATING",
     "162": "NON_PARTICIPATING",
     "163": "NON_PARTICIPATING",
@@ -84,7 +84,7 @@
     "17": "SUCCESS",
     "170": "NON_PARTICIPATING",
     "171": "NON_PARTICIPATING",
-    "172": "SUCCESS",
+    "172": "NON_PARTICIPATING",
     "173": "NON_PARTICIPATING",
     "174": "NON_PARTICIPATING",
     "175": "SUCCESS",
@@ -97,16 +97,16 @@
     "181": "NON_PARTICIPATING",
     "182": "NON_PARTICIPATING",
     "183": "NON_PARTICIPATING",
-    "184": "NON_PARTICIPATING",
+    "184": "SUCCESS",
     "185": "NON_PARTICIPATING",
-    "186": "SUCCESS",
+    "186": "NON_PARTICIPATING",
     "187": "NON_PARTICIPATING",
     "188": "NON_PARTICIPATING",
     "189": "NON_PARTICIPATING",
     "19": "NON_PARTICIPATING",
     "190": "NON_PARTICIPATING",
     "191": "NON_PARTICIPATING",
-    "192": "SUCCESS",
+    "192": "NON_PARTICIPATING",
     "193": "NON_PARTICIPATING",
     "194": "SUCCESS",
     "195": "SUCCESS",
@@ -208,7 +208,7 @@
     "51": "SUCCESS",
     "52": "NON_PARTICIPATING",
     "53": "SUCCESS",
-    "54": "SUCCESS",
+    "54": "NON_PARTICIPATING",
     "55": "SUCCESS",
     "56": "SUCCESS",
     "57": "NON_PARTICIPATING",
@@ -218,7 +218,7 @@
     "60": "SUCCESS",
     "61": "NON_PARTICIPATING",
     "62": "SUCCESS",
-    "63": "NON_PARTICIPATING",
+    "63": "SUCCESS",
     "64": "NON_PARTICIPATING",
     "65": "NON_PARTICIPATING",
     "66": "NON_PARTICIPATING",
@@ -254,7 +254,7 @@
     "93": "NON_PARTICIPATING",
     "94": "NON_PARTICIPATING",
     "95": "SUCCESS",
-    "96": "NON_PARTICIPATING",
+    "96": "SUCCESS",
     "97": "NON_PARTICIPATING",
     "98": "NON_PARTICIPATING",
     "99": "SUCCESS"
@@ -268,20 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5489987,
-    5489991,
-    5489996,
-    5490000,
-    5490003
+    5490071,
+    5490138,
+    5490142,
+    5490146,
+    5490150
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 77,
+  "inner_step": 0,
   "inner_steps": 0,
-  "last_allreduce_block": 5487765,
+  "last_allreduce_block": 5490074,
   "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 sha256:7f27157348c0fccc69e9112f8a498b939234e8b0a26574b9263d6b2adfbe1091
+oid sha256:3c1187b6fd4a15fdb7e0d7205e470d360d291e5680e4a17cac88e01b34083242
 size 8081782503

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0df88b051ce4e4c2cb94df8323894cb32775153e99046c9341d8cbf0337d0428
+oid sha256:feeff45d1021a7e9ff5bba2e352dbf923b3e2978877f5c2c450c6c4f9fdce4d5
 size 4040701744