Run 4. Outer Step 4. Inner Step 16.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486853,
-    5486856,
-    5486859,
-    5486862,
-    5486865,
-    5486868
+    5486871,
+    5486873,
+    5486875,
+    5486877,
+    5486878
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 15,
+  "inner_step": 16,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c517b5a726cca6f55877fe392e297cab49a0be54b8b2bd8166846fd64e1d74fe
+oid sha256:fd92376a4fda31b6e3615171d7e3aceea2189c70fe0551e2eb12c636a2af416f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc92b8cfc4c1e60df168688280753fe4f489fb86dc5398cbeefe2c9fb5d8b47b
+oid sha256:8b309bdde510761b795a16c45279ca54f00de2e7d56727d2af904f2a34dd756a
 size 4040701744