Run 4. Outer Step 29. Inner Step 3.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5540242,
-    5540247,
-    5540252,
-    5540256,
-    5540261
+    5540266,
+    5540270,
+    5540275,
+    5540280,
+    5540284
   ],
   "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": 5539810,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d5096fd17dfd0642913f26559bd59365845f96ae60b3100253fe4b32e0bd72f
+oid sha256:8df6b25c982b729690427aec63e2d2b6689b2652474dd65344b5bdffd76703f4
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8290303ac29ccfee2f98065a413cf0aae57b862b5ce83d6a6977b22599bfd270
+oid sha256:9f0648dc40f599da23d239b82434eb5a7d278ccdf74bc3a03827e5f1bc101998
 size 4040701744