Outer Step 1. Inner Step 12. Batch Size 372

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "distributed/optimized-gpt2-1b",
+  "_name_or_path": "philivy/81",
   "activation_function": "gelu_new",
   "all_reduce_scores": {
     "0": "SUCCESS",
@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5160862,
-    5160872,
-    5160889,
-    5160904,
-    5160918,
-    5160926,
-    5160936,
-    5160951,
-    5160972,
-    5160982
+    5161128,
+    5161157,
+    5161167,
+    5161175,
+    5161186,
+    5161200,
+    5161208,
+    5161232,
+    5161257,
+    5161267
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 7,
+  "inner_step": 12,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:981045c30712f6879cfa55ff2f0a79adee0f94f6a271162e0ef6ee22f3cb0385
+oid sha256:2588844125b9c5f80bda72364644c1973a10cc65a34e6805582b237779fd1a8d
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63ae3560a02fb9c73194651e633841f8b6e465a70e4273ebcb1503f49a064d35
+oid sha256:428793d3aa840bd7a85a27a93112433d6f2ff7c32a79f409b08130bd84e4ea50
 size 4040701744