Outer Step 8. Inner Step 168. Batch Size 84

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5292378,
-    5292381,
-    5292385,
-    5292389,
-    5292392,
-    5292395,
-    5292399,
-    5292402,
-    5292406,
-    5292410
+    5292448,
+    5292452,
+    5292455,
+    5292459,
+    5292462,
+    5292465,
+    5292469,
+    5292472,
+    5292476,
+    5292479
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 157,
+  "inner_step": 168,
   "inner_steps": 0,
   "last_allreduce_block": 5283755,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f69bbfb18e3e33cbf9106b2861c16cb7e03a1a38444b8288ad5cb13e51865e80
+oid sha256:0ab121f87db2a58c32440c2d27751c2d8fe07a276faaf45cfb18377be65553ae
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5eb0fb3d0d812a6885b4e74c5c151aee1fe3091b063c9618a97046f677db5a64
+oid sha256:07acca8d82e538c278522d4a2674f7d087595fd460e2dfb024c171270ebc4bd4
 size 4040701744