Outer Step 0. Inner Step 6. Batch Size 188

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": [
+    5130096,
+    5130106,
+    5130111,
+    5130115,
+    5130125,
+    5130138,
+    5130147,
+    5130163,
+    5130174,
+    5130185
+  ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 6,
   "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:f8a8f53e01ccf912831002cdfca7f142bbfebc985d125c11cfbf550b81ed10ad
+size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:10024d50c5147f2917b8388f2a3aa9528b26da292a533206efbd53083678ed8f
+oid sha256:028c6c298e7cf0934726c8376e98b5f35143c8aa2d8b2b19adb60c0d5ce449ea
 size 4040701744