Run 4. Outer Step 28. Inner Step 27.

config.json

@@ -268,18 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537878,
-    5537880,
-    5537884,
-    5537886,
-    5537888
+    5537890,
+    5537893,
+    5537895,
+    5537896,
+    5537898,
+    5537900,
+    5537902,
+    5537904,
+    5537907,
+    5537910
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 26,
+  "inner_step": 27,
   "inner_steps": 0,
   "last_allreduce_block": 5537093,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f09a5b242059ae5e645cee9032a1951a07c2d78aeed8a6b5a6d116f30e3fce9e
+oid sha256:afcfa617ca9f6276d10724c077a9e67f03cc83f085064d0fdbf9cf53be0e21ef
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9429e3731441dea1c7c4a3f5b2d1ca49fd170357d6898a2c6d16819f6bd6715
+oid sha256:583cc159faeb42bef7c8e17caae3c72f638cd7f1b5077a2500ff0919c299e668
 size 4040701744