Saving weights and logs of step 178261

config.json

@@ -65,7 +65,7 @@
     }
   },
   "tokenizer_class": "GPT2Tokenizer",
-  "transformers_version": "4.9.0.dev0",
+  "transformers_version": "4.11.0.dev0",
   "use_cache": true,
   "vocab_size": 50264,
   "window_size": 256

events.out.tfevents.1632833898.t1v-n-4e27a527-w-0.4170299.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c46cccaa0203135950e5b71efa5b7f9503464a9a08576acee78195f39d534ce
+size 7417090

events.out.tfevents.1633002663.t1v-n-4e27a527-w-0.108588.0.v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0cee2b781287f4a7de1a9abb08b904154063295cec5deda3c2a92302becf926
+size 22317230

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:500f47d5b54bc91a1422319ac39d0cff8404ab2376343ab8d00e11afa3773ab4
+oid sha256:5b90c9c809494625d737d5745a2996027c90a772e5d1d47773e008de96dbfd20
 size 5262371934

run.sh

@@ -7,7 +7,7 @@ python run_clm_mp.py \
        	--do_eval  \
      	--block_size 1024 \
    	--num_train_epochs 5 \
-   	--learning_rate 2e-5 \
+   	--learning_rate 1e-5 \
 	--per_device_train_batch_size 2 \
        	--per_device_eval_batch_size 2 \
     	--overwrite_output_dir \