Outer Step 0. Inner Step 172. Batch Size 112

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5156269,
-    5156273,
-    5156278,
-    5156282,
-    5156288,
-    5156293,
-    5156298,
-    5156303,
-    5156307,
-    5156313
+    5156352,
+    5156357,
+    5156362,
+    5156367,
+    5156372,
+    5156378,
+    5156382,
+    5156387,
+    5156392,
+    5156398
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 162,
+  "inner_step": 172,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8cee6c626b3ce741e9e518da886349771412cd3b995a5f7610ce12fe21c293b
+oid sha256:73d49e8f3d4be43b48c9701ea93e197f097fa72270757bdc375e15f527b9e496
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c43974b4d3ea93a7d6f914323f2830ccd2a11bdf2d88405465ff7449adca2281
+oid sha256:93200775cf737d6f979e7297b3fcee562de0e72b805fc6c1f90109b3a01fdd7a
 size 4040701744