Run 4. Outer Step 28. Inner Step 22.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537657,
-    5537661,
-    5537665,
-    5537669,
-    5537674,
-    5537678
+    5537683,
+    5537687,
+    5537691,
+    5537695,
+    5537699,
+    5537703
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 21,
+  "inner_step": 22,
   "inner_steps": 0,
   "last_allreduce_block": 5531651,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29f20695b5d19439db82f54dca01a2247722de945a31493423d9b2714dd62398
+oid sha256:0bc54c52a12ce491d80f0d7db8c1dd6c10db1c66ba166e03ad132d6cd401562c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c6b15051d60e0adf6938b932d3b8fe659e7f162284b9a7d668b8864f2a48aea9
+oid sha256:0c21b5201691d804f52c6a4263039e916057324316e3b037acb9810e8619014f
 size 4040701744