Outer Step 0. Inner Step 163. Batch Size 284

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5156174,
-    5156187,
-    5156194,
-    5156202,
-    5156206,
-    5156212,
-    5156217,
-    5156221,
-    5156227,
-    5156232
+    5156265,
+    5156270,
+    5156275,
+    5156279,
+    5156284,
+    5156289,
+    5156295,
+    5156300,
+    5156305,
+    5156311
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 154,
+  "inner_step": 163,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09eea8d8ddaba7c1755c1febac18d5c8b91ebde271e97d17a00e5302e70ad7f3
+oid sha256:d82d328253a1b2011e1a59a1b8c4cec35c09838f22e5f4fdba558970af23b559
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b687d94b0650680128e0d6e2bc5e0a35a5632300d13c419f77296e21bf2b4c12
+oid sha256:a6250fa9a770f60359738f9e0d21cc1d36774a1e2e08a618bcde3011ed648f8d
 size 4040701744