Run 4. Outer Step 4. Inner Step 2.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486398,
-    5486405,
-    5486407,
-    5486410,
-    5486413,
-    5486416,
-    5486419
+    5486421,
+    5486424,
+    5486427,
+    5486429,
+    5486431
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1,
+  "inner_step": 2,
   "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 sha256:2f28b899db630d259a0ed22e268b53b008df2dd4d76ce55d83b9c293387ac917
+oid sha256:36b406aea9fd9494b4eb074a7e8d79e6307b70d84fdc415737a89755bb318ee6
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d0d0fda8bc59d78948783cb4e29c0982cdbc5dd37134bdc7d00ec877bda2a8a
+oid sha256:b63a108c603b3bddeec3505d2e89b42a45378c584b37864be0a31385833c8679
 size 4040701744