Outer Step 0. Inner Step 79. Batch Size 148

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5140069,
-    5140073,
-    5140078,
-    5140083,
-    5140087,
-    5140092,
-    5140099,
-    5140106,
-    5140110,
-    5140115
+    5140149,
+    5140153,
+    5140159,
+    5140164,
+    5140169,
+    5140174,
+    5140178,
+    5140183,
+    5140187,
+    5140192
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 70,
+  "inner_step": 79,
   "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 sha256:0d736a17304cfecbb369a5af982dae1c99345fb9da5fbaed2bc82487d13a2466
+oid sha256:2e26ed85cb84730b223607fd37718b2fc2e9342f8488051e8e4747fd20a64942
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97f47959bd40ca80d9c16557bc62cbc5e0fa0dc801d6240a2bf7dbd047b203f7
+oid sha256:9908ff1d8254dc3e8153b6cdc38bec9b20c805973fdae97c5539c900ccf68e82
 size 4040701744