Outer Step 8. Inner Step 100. Batch Size 272

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5291930,
-    5291934,
-    5291937,
-    5291941,
-    5291944,
-    5291947,
-    5291951,
-    5291954,
-    5291958,
-    5291962
+    5292001,
+    5292004,
+    5292007,
+    5292011,
+    5292014,
+    5292018,
+    5292021,
+    5292025,
+    5292028,
+    5292032
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 89,
+  "inner_step": 100,
   "inner_steps": 0,
   "last_allreduce_block": 5283755,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bbd76c7bbf48ceecb3700eee5c3f4d974afd4f5fdf9d2226ba4c4f58a678800
+oid sha256:f96fbdd99d81f142e6f61304ebd9384b7fde57ad320dcbd375380db28d2a14c7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:293ce37f0bc8c3f900961e151e7d139a91c11a8219a3c01e7efd1da2e88750d8
+oid sha256:6fef073a59d72dbb03737d798c75371c6da536957ab27938b617c463d75bae90
 size 4040701744