Run 4. Outer Step 60. Inner Step 6.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5607897,
-    5607901,
-    5607905,
-    5607911,
-    5607915
+    5607920,
+    5607925,
+    5607929,
+    5607933,
+    5607937
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 5,
+  "inner_step": 6,
   "inner_steps": 0,
   "last_allreduce_block": 5607244,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69fb332deafb0d325ca1e27fdd566e4f07708b6e785b393789ea28dcc1971c08
+oid sha256:f74ede4eee4453c037df0dc94f2b7628ebbaadbd3a24723660564fd48331f721
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d43d31cd799b74110638c066d6a9a27b43febbeefba5f73a3b87da83685ed8c7
+oid sha256:5770119f1eb0c3382ae205e4147d1ab5671ce8592bc9e2dc305a44a7bf244c69
 size 4040701744