Run 4. Outer Step 4. Inner Step 21.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487015,
-    5487018,
-    5487021,
-    5487024,
-    5487026,
-    5487029,
-    5487030,
-    5487032
+    5487034,
+    5487037,
+    5487040,
+    5487042,
+    5487045
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "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:c202ab565fb9badb4d2e4c5670d38524584fa01b09aee27b02ccf62128d748ca
+oid sha256:a26e1b5f5e209bc135887065b219f7c76fab189f908a95a675fd889eb1cc2498
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e45d6f1d295be7833ad131c747d3276e38192fc76d47d972168d5ba61d1c028b
+oid sha256:0d2a143f59cb53e21c73f49cced541853eaea867b8b6c17c7d38b72e517c0c55
 size 4040701744