Run 4. Outer Step 26. Inner Step 36.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5535736,
-    5535739,
-    5535742,
-    5535744,
-    5535747
+    5535749,
+    5535752,
+    5535754,
+    5535758,
+    5535760,
+    5535762,
+    5535763,
+    5535765,
+    5535768,
+    5535769,
+    5535771
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 35,
+  "inner_step": 36,
   "inner_steps": 0,
   "last_allreduce_block": 5534554,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d94b93d3b5e4bed7081a78cd818985a1ec6c9be3c790b8927767b62c46870c47
+oid sha256:7f426cbd0d3ecd23f6d3b3c3f549200b38f88b35ed2e9fe5c48fdfe0d31f2320
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4aed56af913f7945de43b62020b5b33b9b7d51f8da09e3be2fc26dd86596229
+oid sha256:4f1b2c6df4ec16983b6abcbb71c18bf9726623c5749115703494d94480921777
 size 4040701744