Run 4. Outer Step 12. Inner Step 2.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5508495,
-    5508499,
-    5508503,
-    5508506,
-    5508510,
-    5508513
+    5508517,
+    5508521,
+    5508525,
+    5508529,
+    5508532
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1,
+  "inner_step": 2,
   "inner_steps": 0,
   "last_allreduce_block": 5506953,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7829bbfdf73968c520b416bf7b02a2c47b936e1d6153475023bc8d744d5ebba0
+oid sha256:1d140bef646993ba3bd38a9d6a1bc6afd52e8ea312acfa0363d6e94c8a4c600c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2cf9149ff01b37536d8fb21eac1090a20fbdeaa7f3e70f4f4492de52dae1e58
+oid sha256:839b770e8fe3c7ca2a41212f487a4611b3f2707a4c9eec4c725dae41f51e34b9
 size 4040701744