Run 3. Outer Step 6. Inner Step 2.

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": [
+    5384727,
+    5384731,
+    5384734,
+    5384738,
+    5384742
+  ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 2,
   "inner_steps": 0,
   "last_allreduce_block": 5378990,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91074fcebac8cb9e44af7f0c432ab1d75be2b0c6f2edde6db4db7e726759416f
+oid sha256:aa529ec7672f3ba11778840332b2cabb7a43f218894c0ed195c5942bb54280bf
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca0236b46615f797360b088a3bcfe5dd20a85f99747917208b8843811f8bd5b4
+oid sha256:c9e384e50c8ce31f97f8cf979853998e20637821fce28f1ff72e61a8f4f3a135
 size 4040701744