Run 4. Outer Step 50. Inner Step 12.

config.json

@@ -268,33 +268,35 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5571796,
-    5571798,
-    5571799,
-    5571800,
-    5571801,
-    5571802,
-    5571803,
-    5571804,
-    5571805,
-    5571806,
-    5571807,
-    5571807,
-    5571808,
-    5571809,
-    5571810,
-    5571811,
-    5571812,
-    5571813,
-    5571814,
-    5571816
+    5571817,
+    5571818,
+    5571819,
+    5571821,
+    5571822,
+    5571823,
+    5571824,
+    5571825,
+    5571826,
+    5571827,
+    5571828,
+    5571828,
+    5571829,
+    5571830,
+    5571831,
+    5571832,
+    5571833,
+    5571834,
+    5571836,
+    5571837,
+    5571837,
+    5571838
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 10,
+  "inner_step": 12,
   "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:fcd24ed4dbb4dc60c51b4b8efa2379dddc5ad5060a829a9fedf33fd743e9bff5
+oid sha256:e287ca02a5c498a0b4f44d7495139d3ebf0ab3c99663c7eac0a6acfa47d5c4ad
 size 8081782503

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00ce6e71d77eadb5929d9622c3181d463ce15f928dd7d21c8b09322fa9d9d05b
+oid sha256:86d026eadc43cdf13726792698e03d944df64a36aa23a12d0e98f0988090f319
 size 4040701744