Outer Step 8. Inner Step 327. Batch Size 488

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5293513,
-    5293514,
-    5293519,
-    5293524,
-    5293525,
-    5293531,
-    5293531,
-    5293536,
-    5293541,
-    5293542
+    5293579,
+    5293585,
+    5293590,
+    5293594,
+    5293596,
+    5293602,
+    5293602,
+    5293608,
+    5293612,
+    5293614
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 316,
+  "inner_step": 327,
   "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 sha256:290a857c61586984d9c3128bc96ab47bd7a87fa1d43ef42df6921cd7ee8ca6ba
+oid sha256:0c758d6700a7100a55df482ade158a89d3350ab38595d84ea8eef311961c7a17
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2b092c51da9033b4c2ca6b95e42cb72002cafbb51ac2a5a3e0eb38692bf71a8
+oid sha256:2e3ea253a144a11bb280296b3425b4605f88ee9396b35c2f801b91001588124a
 size 4040701744