added pytorch model

config.json

@@ -1,4 +1,5 @@
 {
+  "_name_or_path": "./",
   "architectures": [
     "RobertaForMaskedLM"
   ],
@@ -18,6 +19,7 @@
   "num_hidden_layers": 24,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
   "transformers_version": "4.9.0.dev0",
   "type_vocab_size": 1,
   "use_cache": true,

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd182447bab9a7a5f4384ce657f6af7b28ecb0ac3e7e7e57b3a23f5b362e1879
-size 711588089
+oid sha256:daa01e29425ff0d35c30005a143e3832295136dd11c19885af9a8778bcede272
+size 1421662309

flax_model_to_pytorch.py

@@ -4,6 +4,8 @@ import numpy as np
 import jax
 import jax.numpy as jnp
 
+jax.config.update('jax_platform_name', 'cpu')
+
 MODEL_PATH = "./"
 
 model = FlaxRobertaForMaskedLM.from_pretrained(MODEL_PATH)
@@ -14,7 +16,7 @@ def to_f32(t):
 model.params = to_f32(model.params)
 model.save_pretrained(MODEL_PATH)
 
-pt_model = RobertaForMaskedLM.from_pretrained(MODEL_PATH, from_flax=True)
+pt_model = RobertaForMaskedLM.from_pretrained(MODEL_PATH, from_flax=True).to('cpu')
 
 input_ids = np.asarray(2 * [128 * [0]], dtype=np.int32)
 input_ids_pt = torch.tensor(input_ids)

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:853b48bbe067f4afcee874db6e355750e366b46c41549eac7a61e7d54046c6aa
+size 1421780139

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "./", "tokenizer_class": "RobertaTokenizer"}