Outer Step 0. Inner Step 76. Batch Size 500

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5139930,
-    5139938,
-    5139945,
-    5139950,
-    5139955,
-    5139960,
-    5139966,
-    5139979,
-    5139987,
-    5139999
+    5140022,
+    5140025,
+    5140029,
+    5140033,
+    5140038,
+    5140042,
+    5140046,
+    5140050,
+    5140053,
+    5140057
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 69,
+  "inner_step": 76,
   "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:ae48dd564a5a83bca132928fc860dacbcfa126dec03a3eaef3b79411d150565e
+oid sha256:29e00509504999ac1fc72f25ad6ef69caceffd331c352b34f89e0aef685ce293
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23cb9dd7087108c660cde2e80e81c5410afb83b6debfc14d10af308f31877a83
+oid sha256:f598a2e0a2d4112d5c9db225b083759681ffbf031410589d30fb2dfc6b158fa6
 size 4040701744