Run 2. Outer Step 9. Inner Step 18.

config.json

@@ -267,13 +267,44 @@
     "AutoConfig": "distributed/optimized-gpt2-500m--configuration_gpt_optimized.GPTOptimConfig",
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
-  "block_list": [],
+  "block_list": [
+    5356332,
+    5356336,
+    5356340,
+    5356345,
+    5356349,
+    5356354,
+    5356358,
+    5356362,
+    5356367,
+    5356371,
+    5356375,
+    5356379,
+    5356384,
+    5356389,
+    5356393,
+    5356398,
+    5356402,
+    5356406,
+    5356411,
+    5356416,
+    5356420,
+    5356425,
+    5356430,
+    5356434,
+    5356439,
+    5356443,
+    5356448,
+    5356452,
+    5356456,
+    5356460
+  ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 18,
   "inner_steps": 0,
   "last_allreduce_block": 5351170,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50852309c11bba1c68b1fc599de303950b01245d3f8e70c6ec061ebf1229b0c3
-size 2944
+oid sha256:cbc4a42fe77dcac696b441d684738946039a7748ce1dcee4f83c581123c5a313
+size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30a3c2a1f583992135596ada96883cba467a485b6ef53f21ba48c2355f414587
+oid sha256:86a60136b74a72b3b523fa7d562a59e525cc573ead9818609c1ab5aaf3ecade8
 size 4040701744