Run 4. Outer Step 20. Inner Step 20.

config.json

@@ -268,28 +268,27 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525225,
-    5525227,
-    5525228,
-    5525230,
-    5525231,
-    5525232,
-    5525233,
-    5525233,
-    5525235,
-    5525236,
-    5525237,
-    5525239,
-    5525241,
-    5525243,
-    5525245
+    5525247,
+    5525248,
+    5525250,
+    5525251,
+    5525253,
+    5525254,
+    5525256,
+    5525257,
+    5525259,
+    5525260,
+    5525262,
+    5525265,
+    5525266,
+    5525268
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 19,
+  "inner_step": 20,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2af6f97b396dda87209e407dc0f71e49cd1732c86288981ae40512b6764da5ba
+oid sha256:9cc368a135a2f6dc87644a60a4030dabebe32d64e4e5746fd4c5778d7e295dd0
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a27a0a07a271c36d247fe15aaedf40207e196a04850d3fb6c0f0de3a11a3c9e
+oid sha256:6ef605252a30d528234c33e4ae308ec125094cc6ad770f7ac0ae21676375b5fb
 size 4040701744