Run 4. Outer Step 25. Inner Step 40.

config.json

@@ -268,18 +268,22 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5534557,
-    5534560,
-    5534563,
-    5534566,
-    5534569
+    5534573,
+    5534576,
+    5534578,
+    5534581,
+    5534584,
+    5534586,
+    5534588,
+    5534590,
+    5534592
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 39,
+  "inner_step": 40,
   "inner_steps": 0,
   "last_allreduce_block": 5533305,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41c0658ada1b71863b0fa12d44a6ce8592e0b1b8258780211ba4631240bfe858
+oid sha256:c3415b778f161af99bab94402422fbd36311cdb8fe135738d6acc8e7012907b2
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9bb00ebbf85c420194e7ec875dafa2c95ed1e80e61f036657dfa2864b2ce2e0
+oid sha256:075f2c325e60630e1e25309789f92fadb33aad9ec7fceed03abb2ba65cf2ac0b
 size 4040701744