Run 4. Outer Step 50. Inner Step 6.

config.json

@@ -268,31 +268,37 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5571740,
-    5571742,
-    5571744,
-    5571745,
-    5571746,
-    5571747,
-    5571749,
-    5571750,
-    5571751,
-    5571752,
-    5571753,
-    5571754,
-    5571756,
-    5571757,
-    5571759,
-    5571760,
-    5571762,
-    5571764
+    5571766,
+    5571769,
+    5571770,
+    5571772,
+    5571774,
+    5571776,
+    5571777,
+    5571778,
+    5571779,
+    5571780,
+    5571780,
+    5571784,
+    5571785,
+    5571786,
+    5571787,
+    5571789,
+    5571790,
+    5571791,
+    5571792,
+    5571793,
+    5571795,
+    5571796,
+    5571798,
+    5571799
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 4,
+  "inner_step": 6,
   "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 sha256:7373f91dea4aa00e9461da8de72dc07e02a48a3efcbd86bba51cfc12d39af0ad
+oid sha256:035e05de8248e3506a3a990bab0697a878248b6b1decf82250d875b43144c75a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1651134325982f6302b947a402f038a26f6b808c9d67f7ee46dec2aa33fc00f
+oid sha256:0201746c25a1c774f3e26c721bfa0afa0baab075220c512eed79cbcc1a737321
 size 4040701744