Outer Step 0. Inner Step 94. Batch Size 428

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5140083,
-    5140086,
-    5140090,
-    5140093,
-    5140099,
-    5140104,
-    5140110,
-    5140114,
-    5140120,
-    5140126
+    5140150,
+    5140154,
+    5140158,
+    5140161,
+    5140165,
+    5140168,
+    5140171,
+    5140174,
+    5140178,
+    5140181
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 86,
+  "inner_step": 94,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7bf26a9af40bca6d7ec3f9d4011ff41867c7fc2b9c69fb967b86b47883eea606
+oid sha256:6af3dd51a28ba6f05997bfca4725072f684ca84fcbda3ab6ab744d67a83be0f8
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:179e1c451372328bcbe8a9018326c79195ae6c4b3ef8270ab8f31019aead8508
+oid sha256:485651178d4004c3ad575ee2320837162a0cc608f8177cd04a073563ee257fc5
 size 4040701744