Run 3. Outer Step 7. Inner Step 26.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5393836,
-    5393841,
-    5393846,
-    5393850,
-    5393854
+    5393885,
+    5393889,
+    5393893,
+    5393898,
+    5393903
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 24,
+  "inner_step": 26,
   "inner_steps": 0,
   "last_allreduce_block": 5392742,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6fea46176a5cfc3a873bf661f724efebfd39ad53a2269767bf02052209140288
+oid sha256:50b2445aa3fa08facaeeeb151c2b4b6749000d12fa10205e8daf4856a177c8a0
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4f9e98cd88e64833e269a80b0d2a5420dad2a2733842e00162bbe94ce37c67d
+oid sha256:b086562722ac4964119bae68da681f01c625616f778fa441852bc0e3afc335a4
 size 4040701744