Outer Step 0. Inner Step 76. Batch Size 480

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "distributed/optimized-gpt2-1b",
+  "_name_or_path": "Saadaslamkhan/M1",
   "activation_function": "gelu_new",
   "all_reduce_scores": {
     "0": "SUCCESS",
@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5050184,
-    5050188,
-    5050193,
-    5050197,
-    5050202
+    5051147,
+    5051152,
+    5051157,
+    5051162,
+    5051166
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 74,
+  "inner_step": 76,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8aea33f6957b7b44aeb2581652af4c5059bcf67acf0cdf7852e5405d959760c0
+oid sha256:249a277997851334f882ceab32788881433408b0484ea2aef93047a1a9cdfeee
 size 4040701744