Run 4. Outer Step 50. Inner Step 6.

config.json

@@ -268,18 +268,17 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5622258,
-    5622262,
-    5622267,
-    5622271,
-    5622276
+    5622281,
+    5622285,
+    5622290,
+    5622294
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 5,
+  "inner_step": 6,
   "inner_steps": 0,
   "last_allreduce_block": 5571611,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:430415dc851791bc4e37b07266cf66d3725602d2b3ef352b9e5b747da2fdd0bb
+oid sha256:38bd551b47530c89b246bb474e79c6e3dcffeb65595c2b3d0ec4d71c6d843ead
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0320e085b8a9c0ccea6aa85915285d4ef1d4b275e2567714e74689b407cb0be4
+oid sha256:b30085c3d982557aeb5c39cbd193099151b7e054e8bc44bd629121a885e09bda
 size 4040701744