Run 4. Outer Step 28. Inner Step 39.

config.json

@@ -268,23 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5538220,
-    5538223,
-    5538226,
-    5538228,
-    5538231,
-    5538233,
-    5538235,
-    5538236,
-    5538238,
-    5538240
+    5538242,
+    5538245,
+    5538247,
+    5538250,
+    5538252
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 38,
+  "inner_step": 39,
   "inner_steps": 0,
   "last_allreduce_block": 5537093,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db21a4b4ec7c8ee40ab32b8821e2a614861b0f5ae27e47467b825430dd137fde
+oid sha256:2abc9e50ab93feea744e922ff153b25c2508cca93f27cd2b4cdca5474f09fb6b
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9207f403d0c88ec6d340610af54c5a3341165e5b9c137d47518749bc778fd963
+oid sha256:de1fd9fa51757c6d169a20732c342a26e2757966f1dda904b1cb6867efb59758
 size 4040701744