Outer Step 8. Inner Step 290. Batch Size 68

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5293254,
-    5293254,
-    5293260,
-    5293265,
-    5293266,
-    5293271,
-    5293276,
-    5293278,
-    5293283,
-    5293284
+    5293343,
+    5293347,
+    5293349,
+    5293355,
+    5293355,
+    5293360,
+    5293365,
+    5293366,
+    5293372,
+    5293372
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 276,
+  "inner_step": 290,
   "inner_steps": 0,
   "last_allreduce_block": 5283755,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aea5fe295d63379a8084682badfe49fdb5735fb048f84cdeb568e055b2b37f21
+oid sha256:d3e11cd04712f9f1459a25ae035b4606f323f826418f94ddc670ec5623f0727e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f2e8f2e768f1a0f32263738dba733bca56ed2d2d93c1bbcb3a6c284f41b4c74
+oid sha256:6a612fa201358cb855eb2a74970c24b273fcf6ed725af596c9ed542bc642e3d5
 size 4040701744