Upload model and tokenizer

README.md

@@ -0,0 +1,20 @@
+---
+language: en
+license: apache-2.0
+---
+
+# realm-cc-news-pretrained-embedder
+
+## Model description
+
+The REALM checkpoint pretrained with CC-News as target corpus and Wikipedia as knowledge corpus, converted from the TF checkpoint provided by Google Language.
+
+The original paper, code, and checkpoints can be found [here](https://github.com/google-research/language/tree/master/language/realm).
+
+## Usage
+
+```python
+from transformers import RealmEmbedder
+
+embedder = RealmEmbedder.from_pretrained("qqaatw/realm-cc-news-pretrained-embedder")
+```

config.json

@@ -0,0 +1,26 @@
+{
+  "architectures": [
+    "RealmEmbedder"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "hidden_act": "gelu_new",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "realm",
+  "num_attention_heads": 12,
+  "num_candidates": 8,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "retriever_proj_size": 128,
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32e688243425e12024da73b5666fd1e2a36647381237335f025549058c960732
+size 534661981

special_tokens_map.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "do_basic_tokenize": true, "never_split": null, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "bert-base-uncased", "tokenizer_class": "RealmTokenizer"}