Outer Step 10. Inner Step 31. Batch Size 488

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5319505,
-    5319507,
-    5319510,
-    5319512,
-    5319515,
-    5319518,
-    5319520,
-    5319523,
-    5319525,
-    5319527
+    5319559,
+    5319561,
+    5319564,
+    5319566,
+    5319568,
+    5319570,
+    5319571,
+    5319574,
+    5319576,
+    5319579
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 24,
+  "inner_step": 31,
   "inner_steps": 0,
   "last_allreduce_block": 5297756,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3eaecbb0aaffe843a566fceb73f6d91277335e3ad5e3720945d7c2f33bf2293e
+oid sha256:eda253a4e8c009cd3365e9324766ebba6d763899c7b1c7d5a9c7a1fb2a4f9512
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a85fc1abe4c6adbc4b1712e5f3fc4ac5a99807b52c01dc1d065308c1653c4dd
+oid sha256:88f7e51e70d173c04ab8fbba66cdafc17659cf1bc884cc5b33c803af90b8729e
 size 4040701744