Training in progress, step 25

config.json

@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.48.1",
-  "use_cache": true,
+  "use_cache": false,
   "use_sliding_window": false,
   "vocab_size": 151936
 }

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42bf5c04dfdccef5a677a48f42b44ce0ee442a66c3a5b5673ab1775a1c68bc96
+oid sha256:7e57fc1b73c5bac72272afca96cb467bfdbfbbb16c1056c73f5c381070cba980
 size 4957560304

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f4c01d0512cd7efed88c8204c0d14d2c60870e5ccbf759d8dae512eec0970b4
+oid sha256:c9ab8466f8b19a9489316054691e4b745c2c5a050404c99f8b9f74e45eeb71d4
 size 1214366696

runs/Feb07_07-40-41_rzhao17-minitest-lwrrc/events.out.tfevents.1738914108.rzhao17-minitest-lwrrc.261.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc4b847484121d99ac84d241b6c89f3e21bd65463efddefc330549cce83caace
+size 12109

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d49402e05718b75b056a4c9a6664dfe6274fd52e5af9bbb9316565c3c64acff
+oid sha256:98ee2f52abdcc01786dc176b0e031a0337295d031da33f97331d7bd2947ba8bd
 size 6840