Upload TFBertForMaskedLM

README.md

@@ -12,11 +12,11 @@ probably proofread and complete it, then remove this comment. -->
 
 # tf_roberta_v1
 
-This model is a fine-tuned version of [distilroberta-base](https://huggingface.co/distilroberta-base) on an unknown dataset.
+This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 2.1033
-- Validation Loss: 1.8383
-- Epoch: 1
+- Train Loss: 3.0890
+- Validation Loss: 3.0774
+- Epoch: 0
 
 ## Model description
 
@@ -42,8 +42,7 @@ The following hyperparameters were used during training:
 
 | Train Loss | Validation Loss | Epoch |
 |:----------:|:---------------:|:-----:|
-| 2.7196     | 2.1459          | 0     |
-| 2.1033     | 1.8383          | 1     |
+| 3.0890     | 3.0774          | 0     |
 
 
 ### Framework versions

config.json

@@ -1,26 +1,25 @@
 {
-  "_name_or_path": "distilroberta-base",
+  "_name_or_path": "bert-base-uncased",
   "architectures": [
-    "RobertaForMaskedLM"
+    "BertForMaskedLM"
   ],
   "attention_probs_dropout_prob": 0.1,
-  "bos_token_id": 0,
   "classifier_dropout": null,
-  "eos_token_id": 2,
+  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
-  "layer_norm_eps": 1e-05,
-  "max_position_embeddings": 514,
-  "model_type": "roberta",
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
   "num_attention_heads": 12,
-  "num_hidden_layers": 6,
-  "pad_token_id": 1,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "transformers_version": "4.27.2",
-  "type_vocab_size": 1,
+  "type_vocab_size": 2,
   "use_cache": true,
-  "vocab_size": 50265
+  "vocab_size": 30522
 }

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f6f4d4419ed7503239bab1ce63aabc37aa4edcf3f08813b0dc14dc5997d0193
-size 484838716
+oid sha256:473beab816282e4b023332ad8ded564cd81dfdacc89a40a9a7ef25610fed5574
+size 533687680