Training in progress, epoch 1

.gitattributes

@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -1,44 +1,29 @@
 {
-  "_name_or_path": "xlnet-base-cased",
+  "_name_or_path": "xlm-roberta-base",
   "architectures": [
-    "XLNetForSequenceClassification"
+    "XLMRobertaForSequenceClassification"
   ],
-  "attn_type": "bi",
-  "bi_data": false,
-  "bos_token_id": 1,
-  "clamp_len": -1,
-  "d_head": 64,
-  "d_inner": 3072,
-  "d_model": 768,
-  "dropout": 0.1,
-  "end_n_top": 5,
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
   "eos_token_id": 2,
-  "ff_activation": "gelu",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
   "initializer_range": 0.02,
-  "layer_norm_eps": 1e-12,
-  "mem_len": null,
-  "model_type": "xlnet",
-  "n_head": 12,
-  "n_layer": 12,
-  "pad_token_id": 5,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
-  "reuse_len": null,
-  "same_length": false,
-  "start_n_top": 5,
-  "summary_activation": "tanh",
-  "summary_last_dropout": 0.1,
-  "summary_type": "last",
-  "summary_use_proj": true,
-  "task_specific_params": {
-    "text-generation": {
-      "do_sample": true,
-      "max_length": 250
-    }
-  },
   "torch_dtype": "float32",
   "transformers_version": "4.26.0",
-  "untie_r": true,
-  "use_mems_eval": true,
-  "use_mems_train": false,
-  "vocab_size": 32000
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4545880bab4346ebdd63552e9a5fa1832aca7104c48bd192675ccdd9b9c1ea87
-size 469311765
+oid sha256:7f1dd799cbf69e0b32150cf972971117a12c704ec77e680f6df9845a22cc2285
+size 1112254133

runs/Feb05_17-38-42_104169da0f5a/1675618842.7000864/events.out.tfevents.1675618842.104169da0f5a.1049.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7c7aa7ef775a230512dc9dd544018644a8a9db0de2cec7af0ef11ffc53b94cc
+size 5622

runs/Feb05_17-38-42_104169da0f5a/events.out.tfevents.1675618842.104169da0f5a.1049.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:267c18dced45b3098c427cf516f6d6f2995482fd45e338a6fa86208b7574ff15
+size 40

special_tokens_map.json

@@ -1,10 +1,6 @@
 {
-  "additional_special_tokens": [
-    "<eop>",
-    "<eod>"
-  ],
   "bos_token": "<s>",
-  "cls_token": "<cls>",
+  "cls_token": "<s>",
   "eos_token": "</s>",
   "mask_token": {
     "content": "<mask>",
@@ -14,6 +10,6 @@
     "single_word": false
   },
   "pad_token": "<pad>",
-  "sep_token": "<sep>",
+  "sep_token": "</s>",
   "unk_token": "<unk>"
 }

tokenizer_config.json

@@ -1,13 +1,7 @@
 {
-  "additional_special_tokens": [
-    "<eop>",
-    "<eod>"
-  ],
   "bos_token": "<s>",
-  "cls_token": "<cls>",
-  "do_lower_case": false,
+  "cls_token": "<s>",
   "eos_token": "</s>",
-  "keep_accents": false,
   "mask_token": {
     "__type": "AddedToken",
     "content": "<mask>",
@@ -16,12 +10,11 @@
     "rstrip": false,
     "single_word": false
   },
-  "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "xlnet-base-cased",
+  "model_max_length": 512,
+  "name_or_path": "xlm-roberta-base",
   "pad_token": "<pad>",
-  "remove_space": true,
-  "sep_token": "<sep>",
+  "sep_token": "</s>",
   "special_tokens_map_file": null,
-  "tokenizer_class": "XLNetTokenizer",
+  "tokenizer_class": "XLMRobertaTokenizer",
   "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1024d0b94c16e7a1f5202741201a3aaf2137718e639462927b79d270b6bded40
+oid sha256:540f7bc15d4c05f325ebcd8a0cab534b9444b46b540a0278f6a7f7cb8eb0ca2e
 size 3451