Run 4. Outer Step 21. Inner Step 26.

config.json

@@ -268,26 +268,25 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525972,
-    5525973,
-    5525975,
-    5525976,
-    5525978,
-    5525979,
-    5525981,
-    5525983,
-    5525985,
-    5525987,
-    5525988,
-    5525989,
-    5525990
+    5525993,
+    5525995,
+    5525997,
+    5525998,
+    5526000,
+    5526001,
+    5526003,
+    5526004,
+    5526006,
+    5526007,
+    5526009,
+    5526011
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 25,
+  "inner_step": 26,
   "inner_steps": 0,
   "last_allreduce_block": 5525369,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d029b658132e5547de0ec30046a49c516844f33ea2e98703d50b5904c19c358
+oid sha256:cd9ab38b5546ad0f2fb78187621132182082037e4c878dd05fedcbe9165a03b9
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cacfa75d4b4f653cbc074ab200316d306ebf276bd66e1c56118081b8ac0eb4a7
+oid sha256:537b5e076d44a2b8a7401f269f0a11045c775cbb5cf0bcc99aed880b2035a574
 size 4040701744