Add files from Mega pretraining on Colab

Includes tokenizer files (from RoBERTa tokenizer), model weights for the LM encoder, and MLM head weights. Downloaded following the training run in [this Colab notebook](https://colab.research.google.com/drive/1qfUO6o5HRdxBblWlw058HVyvaEPhPpH8?usp=sharing) and uploaded directly to HF hub

encoder_weights.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30d6b26ed273524a5ee9d19958e4f32207062017d87d27b48d60dd7cffdea89a
+size 29335986

mlm_head_weights.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c04e4cc9add1f0df1e5651f831f407df8d73e5ea0fe381f631759aa8855a102
+size 25937803

model_args.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2560e9efc41b09ce87084f22722d000d6e043fa9405b162a1bb53e4e245c3caf
+size 537

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "name_or_path": "roberta-base",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "special_tokens_map_file": null,
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}