Run 4. Outer Step 64. Inner Step 256.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5617565,
-    5617568,
-    5617572,
-    5617576,
-    5617578,
-    5617584
+    5617589,
+    5617593,
+    5617597,
+    5617600,
+    5617603
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 255,
+  "inner_step": 256,
   "inner_steps": 0,
   "last_allreduce_block": 5613899,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62c2dd18f99bc1451de883325907ed295d65a89f9673cc719c0b162d56b71fb2
+oid sha256:8e2ff323b02fb976507b55503c136673ae833ba683464a9227be38a5e098415e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01eed1fc387dea151d9e6092e4775eeb03e744f2efac0b45c6678de2d55bb305
+oid sha256:3bb4a2d9d071b5eb2a7b44cd747b63adf0a52b88bba38e7ed7af2a3edfcb3d70
 size 4040701744