Run 4. Outer Step 4. Inner Step 6.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486535,
-    5486537,
-    5486538,
-    5486540,
-    5486543,
-    5486545,
-    5486549,
-    5486551
+    5486554,
+    5486557,
+    5486560,
+    5486562,
+    5486564
   ],
   "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": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:963ffe282443896976caa6666093e2d3517c08e0f94aca1cc387e6cf5d3a75ce
+oid sha256:dd2dffed63f719a50efeaf65cf37ffd482417f2ade8b8ed4172d32097ae6e7f4
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:21e8d872f30c50110b017ae45d197edbc628339b036b907a943af084574fee44
+oid sha256:d2082ea641862ca8ad95440fe3338a39a44f1f093c5c27d4fc97d7ae11904250
 size 4040701744