Run 4. Outer Step 50. Inner Step 6.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5571861,
-    5571865,
-    5571868,
-    5571870,
-    5571873
+    5571877,
+    5571879,
+    5571881,
+    5571884,
+    5571887
   ],
   "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": 5571611,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9edd1e9ee3610b0f0e0e507ee5f26dd82cf6eabed242fd6454c999b2bc91974b
+oid sha256:f47ab766ce84ac8608a443dc6e3dd09095f65493fa8432771660828f6cc01898
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc9883e3ee53b09b921531e581ca9dd6c161be138674c48452e8769a2d5008a6
+oid sha256:4f2d6c67bce7d12ba0f0c297662178b86baca8d5ec871d4af1cb1485fc63ffcd
 size 4040701744