Outer Step 0. Inner Step 137. Batch Size 316

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5131653,
-    5131664,
-    5131680,
-    5131693,
-    5131705,
-    5131714,
-    5131720,
-    5131731,
-    5131738,
-    5131744
+    5131774,
+    5131783,
+    5131791,
+    5131798,
+    5131805,
+    5131813,
+    5131821,
+    5131837,
+    5131848,
+    5131862
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 129,
+  "inner_step": 137,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:633255d8acec7e01e79411c5d3dbaf19f17b71281bde9a96f5569b918a361383
+oid sha256:4266b1aa557fc357242fd47792072596e67638f6b8b3481a1378793cd287ac74
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed59472bbb7c49b2b3c61b7a7eb8021e99f980e49fb8570a21c08d08640c598a
+oid sha256:252075979cc7b5b3a4d9409eb22db74a1c2d9dfbead2f044d2de5531793fc49c
 size 4040701744