Outer Step 1. Inner Step 1628. Batch Size 448

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5184230,
-    5184234,
-    5184238,
-    5184242,
-    5184246,
-    5184251,
-    5184255,
-    5184259,
-    5184263,
-    5184267
+    5184298,
+    5184302,
+    5184307,
+    5184311,
+    5184315,
+    5184319,
+    5184324,
+    5184328,
+    5184332,
+    5184337
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1619,
+  "inner_step": 1628,
   "inner_steps": 0,
   "last_allreduce_block": 5160840,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:487d2f3f4b2a69af29ae417b3cd0f4d376b3033d74e53dc9a989c0a70e77e36e
+oid sha256:ddef3fcb4abe0f8917fc947658111127220f6ca3618fbec5efd405ba864b1a0e
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df8bfa0fc630f31c92e99ab7f7d1f14cc8743c3c0f8b5af9758d36ad9ac97bdc
+oid sha256:19af7385c208ea7ed8cb3ba686893479d1652fe522f1a01fd16eed4f9d04259c
 size 4040701744