Run 3. Outer Step 0. Inner Step 3.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5365640,
-    5365641,
-    5365642,
-    5365643,
-    5365644
+    5365669,
+    5365671,
+    5365673,
+    5365677,
+    5365678
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1,
+  "inner_step": 2,
   "inner_steps": 0,
   "last_allreduce_block": 5363566,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aabea9bd1072d09be25a264f1e2d05c41fc0327b231e4eca43be4ffb6707f12a
+oid sha256:bcd669bb336b3a1445deb54cf416607fe8774270df141510875bd4846c5cac25
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90b3e428e98fa30e20e42438a9da3f27f84d804d2b9741a9c840c0dcdc0d9ae6
+oid sha256:0e542fc3048d5b661b151352a450a177f6c8f66c9fb5ca2645faa9d706bae462
 size 4040701744