Run 4. Outer Step 40. Inner Step 21.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5555931,
-    5555936,
-    5555940,
-    5555944,
-    5555948
+    5555953,
+    5555957,
+    5555961,
+    5555965,
+    5555970
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5555115,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca48a384621eee9198d758538fe29dec0bbb2ef7b4be5bf361aa5ec0ed4bf712
+oid sha256:154e8de227b50bc65100831a653d2fce1fe022f05aa73cea27fc65c5e1af0ec5
 size 8081782503

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6cff8d45ced267d3510191366eebe56128d2a6cc6eedaeaf945d21a7f9ae018e
+oid sha256:859635a5e820683ad2e36a166c6e8796f5058227338723dd836d97e2c40e16a8
 size 4040701744