Outer Step 0. Inner Step 23. Batch Size 384

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5154944,
-    5154949,
-    5154954,
-    5154958,
-    5154962,
-    5154968,
-    5154973,
-    5154978,
-    5154981,
-    5154986
+    5155017,
+    5155022,
+    5155027,
+    5155031,
+    5155035,
+    5155039,
+    5155044,
+    5155048,
+    5155050,
+    5155056
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 15,
+  "inner_step": 23,
   "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:f05870b535d048b68d20abcb880dff919b5e1fac17e9442c426515bdd25a3211
+oid sha256:7adaef9fa447c3bcbc827e2683e864105b3efc53e7e3962282d432c2154b814f
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be20672ea54bbbbfe26e00b231aa1989e559fb9a9955d2ff8a9f6ad5a25233b8
+oid sha256:a7316354d8097086aecac138461df59dabac9b69ad4850d0fdb1eddbbab37296
 size 4040701744