Run 4. Outer Step 11. Inner Step 32.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5507859,
-    5507863,
-    5507866,
-    5507870,
-    5507873,
-    5507877
+    5507881,
+    5507884,
+    5507888,
+    5507892,
+    5507896
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 32,
   "inner_steps": 0,
   "last_allreduce_block": 5506953,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:766655d214a4c8defccf0f55b5cc227454553ef16e27021d5292282d7894783f
+oid sha256:57c9911aa68231fe3fdde781e7106329b999281dcc968f6657bf02a5579775b7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aadfd395b3169f5498c39fd396cf8a5f6688a7c2eda8db3bcfc7a1149f559a2a
+oid sha256:27b0c077424f2c3db494e4d975bd706f44b3d09aadcd81d882d24974ff201581
 size 4040701744