Run 4. Outer Step 35. Inner Step 39.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5549513,
-    5549518,
-    5549523,
-    5549528,
-    5549532
+    5549536,
+    5549541,
+    5549546,
+    5549550,
+    5549555
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 38,
+  "inner_step": 39,
   "inner_steps": 0,
   "last_allreduce_block": 5548219,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e171abf932573976d76a68127bfbc633662498079f42f7d5218d43cde396d0d5
+oid sha256:1fee9f64e9b7aff689cb35f39d79d13704c398c362731051853b2ad19a4f2657
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f947e357e360dc96d742ed3cb645467112a9e89da2ab6cd3ab985351590f645c
+oid sha256:850ef45c02296d7329804510759ff522fbac39eb3adc185b1516776fcb76f1fa
 size 4040701744