Run 4. Outer Step 28. Inner Step 38.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5538187,
-    5538189,
-    5538191,
-    5538193,
-    5538195
+    5538198,
+    5538201,
+    5538204,
+    5538206,
+    5538208,
+    5538209,
+    5538211,
+    5538212,
+    5538213,
+    5538214,
+    5538216
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "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 sha256:394aa488163fe3b7cc763fe6e965237c68a3df2cdc5d753bedb6fb437ac24e24
+oid sha256:db21a4b4ec7c8ee40ab32b8821e2a614861b0f5ae27e47467b825430dd137fde
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95876161df13a42adb129e13c30b0024bed19b735e44eff24c4cc7740ac82882
+oid sha256:9207f403d0c88ec6d340610af54c5a3341165e5b9c137d47518749bc778fd963
 size 4040701744