Outer Step 8. Inner Step 262. Batch Size 60

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5293089,
-    5293094,
-    5293099,
-    5293100,
-    5293106,
-    5293106,
-    5293112,
-    5293116,
-    5293118,
-    5293124
+    5293166,
+    5293170,
+    5293172,
+    5293177,
+    5293178,
+    5293184,
+    5293188,
+    5293191,
+    5293194,
+    5293198
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 250,
+  "inner_step": 262,
   "inner_steps": 0,
   "last_allreduce_block": 5283755,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a67d1c71d4d2cd47d05c48e1a64b590d9369720328f0da60575339e583c8da87
+oid sha256:ea5f771d03cc41d6b86b37e239ca29a7dc8a87b5b18f8388ec8d8c24884e6a22
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2eab3ec4582e42b8d7efe7e558fe0281f15e6cc39fa86f2636d2e75f0dcd75c3
+oid sha256:08fc50a1e114c378ce4d015585b92d9e0d017af89095460632569ccee3363ba8
 size 4040701744