Outer Step 2. Inner Step 6125. Batch Size 364

config.json

@@ -268,18 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5129010,
-    5129024,
-    5129033,
-    5129041,
-    5129050
+    5129146,
+    5129154,
+    5129165,
+    5129172,
+    5129178,
+    5129187,
+    5129192,
+    5129208,
+    5129217,
+    5129223
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 6120,
+  "inner_step": 6125,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e2b9158302daefbdfbc4610ce95673c7983b85c6f708080f7d855bb8fbca6dc
+size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3eb0bc75751293ece40a989d12ed842b89601117669befa4911e463af465550c
+oid sha256:2f8dc6b7692ea323ebb02c6e881e3075104b1465f304c877d749e9b47bc6d93a
 size 4040701744