Outer Step 10. Inner Step 250. Batch Size 424

config.json

@@ -268,23 +268,25 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5321241,
-    5321245,
-    5321249,
-    5321253,
-    5321257,
-    5321261,
-    5321265,
-    5321269,
-    5321274,
-    5321278
+    5321309,
+    5321313,
+    5321317,
+    5321321,
+    5321326,
+    5321330,
+    5321334,
+    5321339,
+    5321343,
+    5321347,
+    5321353,
+    5321364
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 239,
+  "inner_step": 250,
   "inner_steps": 0,
   "last_allreduce_block": 5265017,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:281cc652dcf13fe1ff23555bba98cf3c1d58aa58f47ad6c7617c46870f1d2155
+oid sha256:25d9f9bb0ba3e47d151aa4ea8c6d98bca5d37eed103ede4838d6d476dc647b17
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:458c5ea64fc62bd448d0cc546a2cb58306efb8a78b73a849f8082cbfc04e947c
+oid sha256:eed86304c94c2e2333c31d0ed9d8f8e32f283bd585d1a6965a516fe40476f0b7
 size 4040701744