Run 4. Outer Step 54. Inner Step 53.

config.json

@@ -267,13 +267,19 @@
     "AutoConfig": "distributed/optimized-gpt2-500m--configuration_gpt_optimized.GPTOptimConfig",
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
-  "block_list": [],
+  "block_list": [
+    5595616,
+    5595622,
+    5595626,
+    5595630,
+    5595634
+  ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 52,
+  "inner_step": 53,
   "inner_steps": 0,
   "last_allreduce_block": 5590498,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d195b5a05a6ecc5947b47460b7050a3709335b446e50142c9f4fea935c32501c
+oid sha256:16730dcdb44fd9a9910610fe4a4859a10d125ed64b8f2af7c1a8673de71bee99
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac0e1fc537f84fd7a2324f5a07c3ba9c36449a7c2cbbe1659a046959850b8cbf
+oid sha256:55737f3925291e2d9a982e669bc7956e5b67de5aed424a432e904ad03b0a1ecc
 size 4040701744