Run 4. Outer Step 36. Inner Step 33.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5550527,
-    5550531,
-    5550535,
-    5550539,
-    5550543
+    5550548,
+    5550552,
+    5550556,
+    5550560,
+    5550564
   ],
   "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": 5549491,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7a7e4f4027c0560b34302a9e058a5e77cb4855892c897dbfc63dc2cc11972df
+oid sha256:c288e690a77ee20792d533bd3c94b2e28b5025236bce6707998688fdb3469d54
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d913977d51f930c8fe58402b710ed17e58aa7faba96d4958d40cca0c3e02781
+oid sha256:90468a18a8fec8fa51569a6b6b68e5490ee59a6774b33e76e1884bbd367216d0
 size 4040701744