Improved supervisor with deepset/tinyroberta-6l-768d

config.json

@@ -1,25 +1,30 @@
 {
-  "_name_or_path": "deepset/minilm-uncased-squad2",
+  "_name_or_path": "deepset/tinyroberta-6l-768d",
   "architectures": [
-    "BertForSequenceClassification"
+    "RobertaForSequenceClassification"
   ],
   "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": 384,
+  "hidden_size": 768,
   "initializer_range": 0.02,
-  "intermediate_size": 1536,
-  "layer_norm_eps": 1e-12,
-  "max_position_embeddings": 512,
-  "model_type": "bert",
+  "intermediate_size": 3072,
+  "language": "english",
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "name": "Roberta",
   "num_attention_heads": 12,
-  "num_hidden_layers": 12,
-  "pad_token_id": 0,
+  "num_hidden_layers": 6,
+  "pad_token_id": 1,
   "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
   "transformers_version": "4.24.0",
-  "type_vocab_size": 2,
+  "type_vocab_size": 1,
   "use_cache": true,
-  "vocab_size": 30522
+  "vocab_size": 50265
 }

special_tokens_map.json

@@ -1,7 +1,15 @@
 {
-  "cls_token": "[CLS]",
-  "mask_token": "[MASK]",
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "unk_token": "[UNK]"
+  "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>"
 }

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d7e1c920890824e816b9b19ad41ad9f674ff7dcaa1a4b25349897523120aa9e
-size 133748192
+oid sha256:176c061a588de46a0826c40dcb6e44480de3cc6e5cd337cd4f5f285e0438281f
+size 328629336

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c571f1b1973291370192db62060efdfb564c9a11865c568693ac7420e857cf55
-size 711494
+oid sha256:1f325de61b2d8f48f2580c234ef57f927ffb952ea8617a9820559ec6b7b687e5
+size 2108717

tokenizer_config.json

@@ -1,17 +1,18 @@
 {
-  "cls_token": "[CLS]",
-  "do_basic_tokenize": true,
-  "do_lower_case": true,
-  "full_tokenizer_file": null,
-  "mask_token": "[MASK]",
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "do_lower_case": false,
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
   "max_len": 384,
-  "name_or_path": "deepset/minilm-uncased-squad2",
-  "never_split": null,
-  "pad_token": "[PAD]",
-  "sep_token": "[SEP]",
-  "special_tokens_map_file": "special_tokens_map.json",
-  "strip_accents": null,
-  "tokenize_chinese_chars": true,
-  "tokenizer_class": "BertTokenizer",
-  "unk_token": "[UNK]"
+  "model_max_length": 512,
+  "name_or_path": "deepset/tinyroberta-6l-768d",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "special_tokens_map_file": null,
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
 }

training.json

@@ -1,12 +1,12 @@
 {
     "epochs": 2,
     "learning_rate": 1e-05,
-    "model_name": "deepset/minilm-uncased-squad2",
+    "model_name": "deepset/tinyroberta-6l-768d",
     "evaluation": [
-        0.03355833771390371,
-        0.036146737329307954,
-        0.9521034393207527,
-        0.9619200527368064,
-        0.03936894991757511
+        0.001422734576696505,
+        0.001509308632540951,
+        0.9986003600680058,
+        0.9986481850885693,
+        0.011450733192732265
     ]
 }