Run 4. Outer Step 18. Inner Step 30.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5521947,
-    5521951,
-    5521954,
-    5521958,
-    5521961,
-    5521965
+    5521970,
+    5521974,
+    5521977,
+    5521981,
+    5521985,
+    5521988
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 29,
+  "inner_step": 30,
   "inner_steps": 0,
   "last_allreduce_block": 5509976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fffc44d6f2d65c770fed90400fbf183d72131b059325525dfa20567926cd7b79
+oid sha256:1c6b719e57374cf7e692b901cad0b97e0581965d27a0aecf8dfc225c143cfcb2
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f43b14717cfdcb7b6b1a1398df34bc9d834f9228d88cf89a7cd7f22a1e653bfc
+oid sha256:71ab978ca4bcf084af51376a07a384590a3230f22e3f79ee48b4a6bb4f157503
 size 4040701744