Run 4. Outer Step 35. Inner Step 38.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5549376,
-    5549380,
-    5549384,
-    5549388,
-    5549392
+    5549396,
+    5549401,
+    5549405,
+    5549409,
+    5549413
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5548219,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c2abb86db7535ae333e623b10f4c07e76404908ca8f9578a2bb927518af8f1ef
+oid sha256:61ba0d79d63b98a4ea46cf637ec9ddfa9db540fde14116bfbfae59cbcb15d127
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5868ea4f47242786de977401a09c4bcfeee29428de30d185dcf9db125919bcfc
+oid sha256:a39db3a59bf26864790725343d2218b1b7768dcb110c9913ac234088def386f8
 size 4040701744