Outer Step 8. Inner Step 67. Batch Size 192

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5291709,
-    5291713,
-    5291716,
-    5291719,
-    5291723,
-    5291727,
-    5291730,
-    5291734,
-    5291737,
-    5291741
+    5291783,
+    5291786,
+    5291790,
+    5291793,
+    5291797,
+    5291800,
+    5291804,
+    5291807,
+    5291811,
+    5291814
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 55,
+  "inner_step": 67,
   "inner_steps": 0,
   "last_allreduce_block": 5283755,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c6e409dbe384af6fadb99103b9f66eb5152c1649ddf20cd22c0ca042f30cc68
+oid sha256:540d72b12e00169f032865e58a2955216702c02afd9738c830be8e7bc6f7286d
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6723b689ea74009eaedc1e3b52a5fe1576bcab1ee6529b7bad216e4a55633c46
+oid sha256:23b2b3c7fbf41e072a73a87e92237148076673453a024e60166444971d501b1a
 size 4040701744