Run 4. Outer Step 28. Inner Step 6.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537206,
-    5537210,
-    5537214,
-    5537219,
-    5537223
+    5537228,
+    5537232,
+    5537236,
+    5537240,
+    5537245
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 5,
+  "inner_step": 6,
   "inner_steps": 0,
   "last_allreduce_block": 5535854,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:669d3ec455327ba9e68ecb0f92f56f11cff514cff40c1fab7b6a09be639bcbd2
+oid sha256:8b2f5b1ca07df932823139e2d9547f0c05e3e12c7bbdbb0a82d82c0463b013ac
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8c8b47e01a56fd09ac28e225df000406ea8e82637e7a0969e4b2ea299da0895
+oid sha256:43c92563c95bbf0285058d0faad4529e48fc66fc3fa3d893d59684859e7e5bfd
 size 4040701744