Run 4. Outer Step 62. Inner Step 39.

config.json

@@ -268,9 +268,9 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5611644,
-    5611648,
-    5611653,
+    5611658,
+    5611658,
+    5611658,
     5611658,
     5611658
   ],
@@ -279,7 +279,7 @@
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 38,
+  "inner_step": 39,
   "inner_steps": 0,
   "last_allreduce_block": 5602341,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37c1a86feb98b1e7878c8e604b598402cbd44197fe88c1af537c4c2f76f74b63
+oid sha256:787d9a91b0cf9b046b45538c3dbdbd09351945c21671084acaba9a78c50c8e57
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0d4c769390a473c65cab53da0f5a94cf5ec43e2cd75b100e37294a45b6d17f0
+oid sha256:621e9a441e7b003d8e336e49c1e142e52e9db8361152226f15edcfcc15622d50
 size 4040701744