Training in progress, step 500

config.json

@@ -35,6 +35,6 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 128256
 }

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:420037bfc390c69f2351a649e9d4c19b4248f715188d7b9d8aaf71f557ae2a6d
+oid sha256:a949d0d551d426eda481df55c9ce61d9e9f886cf169e5144df329731c5587075
 size 4965799096

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c07cf39e09b65eec416d46c14632a76e08bf99a27f76c57718c0f8cb7394df7d
+oid sha256:34d352336d2f3d15b0fdace3f3e6e2ba4213d9a93cc7ff41fc3f50a8dd9beca7
 size 1459729952

runs/Sep29_22-34-07_action-graph-trainer/events.out.tfevents.1727649567.action-graph-trainer.718565.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f7e0fcd867fd94d460c98b736ed95a36268de11fbfe7d6d70cef64828224829
+size 45066

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af07ef307fbfb06288cb6621a9393b950ecf353d772cefedc6ac3c3a44a47f9b
+oid sha256:85d6c510bba738cfccc244be84a0a0f2463353b3a8d53114d995188b1e3f1090
 size 7544