* context increase tuning

config.json

@@ -45,7 +45,7 @@
   "sparse_pred_ignore_index": -100,
   "sparse_prediction": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.50.0.dev0",
+  "transformers_version": "4.50.3",
   "unk_token_id": 1,
   "vocab_size": 32000
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd3f7ec29a381d3036f0629df946d25cce6393d965b713cdceb0ad0b95661371
+oid sha256:c04ce76a5434dc05f4c5f71d5fdc192b3680d1e34f9207759a87b45181f99d3f
 size 542135064

tokenizer.json

@@ -2,13 +2,13 @@
   "version": "1.0",
   "truncation": {
     "direction": "Right",
-    "max_length": 1024,
+    "max_length": 8192,
     "strategy": "LongestFirst",
     "stride": 0
   },
   "padding": {
     "strategy": {
-      "Fixed": 1024
+      "Fixed": 8192
     },
     "direction": "Right",
     "pad_to_multiple_of": null,

tokenizer_config.json

@@ -49,6 +49,10 @@
   "mask_token": "[MASK]",
   "max_len": 8192,
   "max_length": 8192,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
   "model_max_length": 8192,
   "never_split": null,
   "pad_to_multiple_of": null,
@@ -59,12 +63,8 @@
   "stride": 0,
   "strip_accents": false,
   "tokenize_chinese_chars": true,
-  "tokenizer_class": "PreTrainedTokenizerFast",
+  "tokenizer_class": "PreTrainedTokenizer",
   "truncation_side": "right",
   "truncation_strategy": "longest_first",
-  "model_input_names": [
-    "input_ids",
-    "attention_mask"
-  ],
   "unk_token": "[UNK]"
 }