Run 4. Outer Step 2. Inner Step 0.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5480893,
-    5480896,
-    5480901,
-    5480905,
-    5480909,
-    5480912,
-    5480916
+    5480960,
+    5480987,
+    5480991,
+    5480995,
+    5480999
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 3,
+  "inner_step": 0,
   "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 sha256:89c702988b00529307d376d124e7283f6657aeb6e6e75fa3f95fbc6b66c4edf6
-size 8081782026
+oid sha256:a417258a25122d504f71d614ffe4538b38330abb16370306e7ea074fd9310445
+size 2944

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b5398afe5ad7a603c127d58e81d41dea31e30dc51c62ccfd94b9d4369b906a1c
+oid sha256:712e598b78200841652fe175a4a9507ccf6781224c0066a2c46b2fc0b2643d29
 size 4040701744