Outer Step 0. Inner Step 6. Batch Size 196

config.json

@@ -267,13 +267,24 @@
     "AutoConfig": "distributed/optimized-gpt2-500m--configuration_gpt_optimized.GPTOptimConfig",
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
-  "block_list": [],
+  "block_list": [
+    5154870,
+    5154875,
+    5154880,
+    5154885,
+    5154889,
+    5154894,
+    5154899,
+    5154903,
+    5154907,
+    5154912
+  ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 6,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91370d8caa12f3d0605c1431c4c8639762dc3d8963d1477ae3c1347a6cf2c93a
-size 2688
+oid sha256:f376947e05645d30abbea1207f4435714d225a160f3bbe5f0c4454fd3f28fda0
+size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcbe9e05268c20b6c5a24174c0482f755f8a38e4487e990bdcc7b98834890d4c
+oid sha256:cbfa703554ece643115da1e286dce21872bb1c42eb0eb0e4b3d71064b1284d16
 size 4040701744