Outer Step 5. Inner Step 652. Batch Size 224

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5276962,
-    5276966,
-    5276970,
-    5276975,
-    5276979,
-    5276983,
-    5276988,
-    5276992,
-    5276996,
-    5277000
+    5277036,
+    5277041,
+    5277045,
+    5277049,
+    5277053,
+    5277058,
+    5277062,
+    5277066,
+    5277071,
+    5277075
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 642,
+  "inner_step": 652,
   "inner_steps": 0,
   "last_allreduce_block": 5272223,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b9a22c3eb1162b6bb2d79ba81826da20bee137bcec0c71ea5cc60a3cc88db1d
+oid sha256:5c2134b2f6ad916bf2ed8451e7cbd555ecc0684caf6d4189a9416414d53a388a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe63b5edca711cc7af9bec4ccd225dd56e9b3414cf93e40abda292b7b51862b0
+oid sha256:ba86dc5655643ab04088a5e1a9b60aadfe8bb2275359501066f74906c19da4a0
 size 4040701744