Run 4. Outer Step 32. Inner Step 33.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543808,
-    5543810,
-    5543813,
-    5543816,
-    5543819,
-    5543822,
-    5543824,
-    5543826
+    5543828,
+    5543830,
+    5543832,
+    5543835,
+    5543837
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 32,
+  "inner_step": 33,
   "inner_steps": 0,
   "last_allreduce_block": 5542700,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:310bbb2ff306f0802c3bbf2420e270d3cbf5d17de45c0471fe829c93fabfa475
+oid sha256:357993512ab8136fc19dd5e78091cd1a745c1ff1773edbb97bed1cc685b71318
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b661c52940d0fb3d153df76b2333a710d0d75609a09423fa154aba716640719
+oid sha256:459ab60d00f6a2abdb01fe8e7ff2bbe8a492d96698db291d2c1bc88e624fe77a
 size 4040701744