[Awaiting approval] Upload ONNX weights

[Automated] Converted using [Optimum](https://github.com/huggingface/optimum). Models will be merged manually by

@Xenova

once they have been checked with [Transformers.js](https://github.com/xenova/transformers.js).

onnx/decoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c64acd984feea67a5ecd8ef26df92899d471a7d009ef16aa684742500958985
+size 3838876

onnx/decoder_model_merged.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95044b774b81a8b23fb2ad7d36c1b0aba6f72d83f1d1b2a3d5de39139480fc7d
+size 3923404

onnx/decoder_with_past_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cd53aaa6807669369f339bb0e817504b00c0b8a7d9cc943895fdfa2f65ff1e5
+size 3815590

onnx/encoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c6025b7e40d02890e5fcc9fd0c0dcdb7cd5f9cac54c4b4313451da3e9c7ec41
+size 3774560