Run 4. Outer Step 28. Inner Step 19.

config.json

@@ -268,18 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537650,
-    5537653,
-    5537656,
-    5537658,
-    5537660
+    5537662,
+    5537663,
+    5537665,
+    5537667,
+    5537669,
+    5537672,
+    5537674,
+    5537677,
+    5537679,
+    5537681
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 18,
+  "inner_step": 19,
   "inner_steps": 0,
   "last_allreduce_block": 5537093,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09b5021626f5a0deee8e992c31a11f6cd7b30601dbb5b51317431eae7ff734ef
+oid sha256:7d1be1b2471d241638a14f75eb12b4f5cfc9f075bf04b05c9d4525ebd3b1b216
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b3388ca2a14fdebeac4434893fe6c7e5f922bd6cb0f72c92ab063cc7ab0e4175
+oid sha256:04470f370ca59a4e30309ccd163d30eaf25f0a17f945174e4c951696a63a3932
 size 4040701744