Run 4. Outer Step 18. Inner Step 3.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5521064,
-    5521068,
-    5521072,
-    5521076,
-    5521079,
-    5521083
+    5521088,
+    5521091,
+    5521095,
+    5521099,
+    5521102,
+    5521106
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2,
+  "inner_step": 3,
   "inner_steps": 0,
   "last_allreduce_block": 5509976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe1a76850038f2262d83346f822e5d164feec7856eb70c3f13916165779b0712
+oid sha256:df711a1b9edf1797caa8a01cf9301955edda90303a06e579cd909208af480697
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8eded399234cb5048075b1439329ef4c3c45789e7bc83c3a7f04508eb4fcb47b
+oid sha256:491363b22bd8a7b3be479778bc515df59f7b30ae2a392b32243e986b0e4d3491
 size 4040701744