Outer Step 2. Inner Step 128. Batch Size 152

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5147764,
-    5147768,
-    5147773,
-    5147777,
-    5147781,
-    5147785,
-    5147789,
-    5147794,
-    5147799,
-    5147802
+    5147838,
+    5147843,
+    5147847,
+    5147851,
+    5147855,
+    5147859,
+    5147863,
+    5147868,
+    5147872,
+    5147876
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 118,
+  "inner_step": 128,
   "inner_steps": 0,
   "last_allreduce_block": 5146835,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6643b528acc62d1d58584ad0b49c6e0de66cc2219ff4e8a6f82319e95d158e84
+oid sha256:16bdd49a2c092c101fbc8270033b9c85ba7a9e0ae5ef837b4e6332dcb0e62a3e
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b592b24f7a2b40c8a3a105125177a271ecba15815344a8461849a4f9309262f2
+oid sha256:c6d66e2d8d69a444a2c1b2cea0e041966c890ff8bd7bab5cbfbe22229472ab6d
 size 4040701744