Outer Step 10. Inner Step 260. Batch Size 268

config.json

@@ -268,25 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5321309,
-    5321313,
-    5321317,
-    5321321,
-    5321326,
-    5321330,
-    5321334,
-    5321339,
-    5321343,
-    5321347,
-    5321353,
-    5321364
+    5321401,
+    5321405,
+    5321409,
+    5321413,
+    5321417,
+    5321421,
+    5321425,
+    5321430,
+    5321433,
+    5321437,
+    5321444
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 250,
+  "inner_step": 260,
   "inner_steps": 0,
   "last_allreduce_block": 5265017,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25d9f9bb0ba3e47d151aa4ea8c6d98bca5d37eed103ede4838d6d476dc647b17
+oid sha256:9f58b27bb9ed5fd8a856b83587694186a0f15ec4a91c44d7ae7803efd0ae9c5f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eed86304c94c2e2333c31d0ed9d8f8e32f283bd585d1a6965a516fe40476f0b7
+oid sha256:c589527005b38651d7e92afe907a9cb376d18f044458374c66e595f965f3b277
 size 4040701744