Outer Step 0. Inner Step 5. Batch Size 380

config.json

@@ -267,13 +267,24 @@
     "AutoConfig": "distributed/optimized-gpt2-500m--configuration_gpt_optimized.GPTOptimConfig",
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
-  "block_list": [],
+  "block_list": [
+    5129845,
+    5129853,
+    5129863,
+    5129869,
+    5129875,
+    5129879,
+    5129884,
+    5129891,
+    5129898,
+    5129906
+  ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 5,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91370d8caa12f3d0605c1431c4c8639762dc3d8963d1477ae3c1347a6cf2c93a
-size 2688
+oid sha256:7b0d644681fbe1e809d25c867c2caa64966e961254f604c436510e6c2307bbe6
+size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:10024d50c5147f2917b8388f2a3aa9528b26da292a533206efbd53083678ed8f
+oid sha256:e16490978e9f32d81e36ff616abec40cd6eb70901c4a798619830a95f271c6b6
 size 4040701744