Run 4. Outer Step 32. Inner Step 38.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5543915,
-    5543920,
-    5543925,
-    5543929,
-    5543934
+    5543939,
+    5543944,
+    5543949,
+    5543954,
+    5543959
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5542700,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de59d85a0f830ec3ad13f39abeeb068c8191b1119e272ef15e8a30a2be874364
+oid sha256:70f8a81387246ef586055fae52f2d34ea382d707b30c19f22e359f88f97ecb90
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c83696832b1c3d483b9038e5d8f1f0026d4e2e39bebdd9074f7e7387725b5df4
+oid sha256:010e92a8e87ea49d9260a4009c8f7a01f67ca533a1d16a3422d4f7fcc2f6eea5
 size 4040701744