Run 4. Outer Step 22. Inner Step 8.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526870,
-    5526876,
-    5526883,
-    5526889,
-    5526894
+    5526902,
+    5526909,
+    5526915,
+    5526921,
+    5526927
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 7,
+  "inner_step": 8,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78d6669bf583a8c91e98b0df8bfaa424508f35b9a56073dab195b7647b3a9043
+oid sha256:57dda6d6866536cd829652df3793b83953fcc6141d18b9027b3c7d8487583272
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97a9ad0d62e7fd9dd2de5633dd7878403f32acef019a9a43795ee07051ac93e1
+oid sha256:af2134734e95147c98b98f4891a48e5d16b730af1b1601cbc5661e0240b8d350
 size 4040701744