Run 4. Outer Step 62. Inner Step 32.

config.json

@@ -268,9 +268,9 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5611647,
-    5611651,
-    5611655,
+    5611658,
+    5611658,
+    5611658,
     5611658,
     5611658
   ],
@@ -279,7 +279,7 @@
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 32,
   "inner_steps": 0,
   "last_allreduce_block": 5610584,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e220253a4815f63c8f6f277c685b1ded9e8775ea9c06db30a650b8fd1631b39
+oid sha256:bb60841fea7513cb8ca9a5846295ba1f18dd26d240f054d35f95898dfdf087a7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f99ad6c6f5f02aefe956893959679707fbb339597a28c1509275c45e4c5891e
+oid sha256:faae64c14c3dd634359770be8e5503536440417be91cedae17b98d672afc9e9b
 size 4040701744