Run 4. Outer Step 2. Inner Step 2.

config.json

@@ -268,18 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5480844,
-    5480848,
-    5480852,
-    5480856,
-    5480860
+    5480865,
+    5480869,
+    5480873,
+    5480877,
+    5480881,
+    5480885,
+    5480889
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1,
+  "inner_step": 2,
   "inner_steps": 0,
   "last_allreduce_block": 5480541,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1ccadb09f732ae9136b4da8478784de01c191f9e442c736aed2365639eb00b1
+oid sha256:eb8cb2d49681b8677a1ff6398e69b8a8079b294c6fef163371aecfe358fcd054
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:712e598b78200841652fe175a4a9507ccf6781224c0066a2c46b2fc0b2643d29
+oid sha256:cc1145a7c7c8b07fe2e416728e2ca92c40c8104585fef90f4a5594265b20baff
 size 4040701744