Run 4. Outer Step 4. Inner Step 19.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486948,
-    5486951,
-    5486954,
-    5486956,
-    5486958,
-    5486960,
-    5486962
+    5486965,
+    5486967,
+    5486970,
+    5486972,
+    5486974
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 18,
+  "inner_step": 19,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4dd16fb7920067ab57d67fd69d8d8468781efbb7df4f56e732b5bc08a21c778d
+oid sha256:8fd480c6e187f18f788567a8c5157d2dbb14a30100627ef3127ab538b67c5bed
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fccb2f48e9a9ca798eecd4af5769f971dd7ef0cf8b9b85f11417785705d036c5
+oid sha256:453b486a6435aac5b1ab2fafe8a6ef7782346061530f04898190dbda9cc3bb3e
 size 4040701744