Outer Step 0. Inner Step 170. Batch Size 16

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5156270,
-    5156274,
-    5156279,
-    5156283,
-    5156289,
-    5156293,
-    5156299,
-    5156304,
-    5156309,
-    5156315
+    5156353,
+    5156357,
+    5156362,
+    5156365,
+    5156371,
+    5156376,
+    5156381,
+    5156385,
+    5156390,
+    5156396
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 160,
+  "inner_step": 170,
   "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 sha256:0a0790e146a1a936974de8c366d4a2eaa1db22d2aafe9be2ec4287f12362cbab
+oid sha256:f300d3dfd37af51ccf109265f4d559c075c8951adab2099a92694cc42167c325
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e80c63a5fa2d624c41422a4ef85fb910b94783473340ec0cbff41b1e0206a6d7
+oid sha256:6bf09aa3909428596439b47228957625c0c96e637ca0ab5167a84b0bbae4f5f2
 size 4040701744