Run 4. Outer Step 38. Inner Step 3.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5553639,
-    5553643,
-    5553647,
-    5553651,
-    5553655
+    5553660,
+    5553664,
+    5553669,
+    5553673,
+    5553677
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2,
+  "inner_step": 3,
   "inner_steps": 0,
   "last_allreduce_block": 5552497,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd64dd60f83de3bba716a5eed294a85f2bd5d8f05393225541c0c2f1f3339bd7
+oid sha256:e054f7ef27156393e6c8746b643411d3a469b706e0a3ae2c31a649726b8970d9
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fc0626ec9f0ee7bd6bb74213933f731b3f80d0db77999aca5569db69709a537
+oid sha256:5093732f0a9a9d13281b72e6a15a7c72c04f3785711f2b54130fc655268b7176
 size 4040701744