Outer Step 0. Inner Step 60. Batch Size 108

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5155242,
-    5155247,
-    5155252,
-    5155257,
-    5155262,
-    5155267,
-    5155272,
-    5155276,
-    5155281,
-    5155285
+    5155319,
+    5155324,
+    5155329,
+    5155334,
+    5155339,
+    5155345,
+    5155349,
+    5155353,
+    5155357,
+    5155361
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 51,
+  "inner_step": 60,
   "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:bc11b825ea85525daaab978f837341377a886e0061e064182733bac007857185
+oid sha256:b614fe9d3601bdd722ef05844b005bb66066152d895f7837c1e8e92414a508a2
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f7bf5a8b9b4aa48987db2732fcbdf058588c43c27111358d8614f959b66c52b
+oid sha256:2704546bf3ce0cca028680acc0a6388cc5bf1375033d1835528632f5fd961ef3
 size 4040701744