Run 4. Outer Step 97. Inner Step 3.

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "distributed/optimized-gpt2-1b",
+  "_name_or_path": "zeras141a/d8",
   "activation_function": "gelu_new",
   "all_reduce_scores": {
     "0": "NON_PARTICIPATING",
@@ -268,15 +268,15 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5691781,
-    5691797
+    5691782,
+    5691802
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 5,
+  "inner_step": 3,
   "inner_steps": 0,
   "last_allreduce_block": 5686307,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c33dac81ef114b545094fa438c2c6cbb48747ecd04a9caf1c4964232753572a
+oid sha256:0030fe7d6fcb420eb1f087734bcdb207ddce6227b502f24bacf3431ba2996d87
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d915c862f087c05189f330e7539db7a13795d62332d8e7ae5da34be20c5ed65e
+oid sha256:fee9bf97993d35d4946a69727e970171f3e841d888f56adbcc35bd583d0b206e
 size 4040701744