Run 4. Outer Step 32. Inner Step 36.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543842,
-    5543847,
-    5543852,
-    5543857,
-    5543861
+    5543867,
+    5543872,
+    5543876,
+    5543881,
+    5543886
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 35,
+  "inner_step": 36,
   "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:bbef61aeb625720ef71f19e2dd7ddcd2a7a1a124d3078fdb9ed5bd88a7f6a7c7
+oid sha256:75bfbaf9f950cec0a62412610e611e5870bd336220ad73301b658f8bc1b5adf8
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa0194528280cf432622c6ee3f1b8d867a22d5705c80eea28088f7b4764177e1
+oid sha256:c3fc11eb920b711f13136873102cfb238e76e70b20cf11c4ffb813dadbba34bb
 size 4040701744