Upload tokenizer

tokenizer.json

@@ -1,7 +1,21 @@
 {
   "version": "1.0",
-  "truncation": null,
-  "padding": null,
+  "truncation": {
+    "direction": "Right",
+    "max_length": 300,
+    "strategy": "LongestFirst",
+    "stride": 0
+  },
+  "padding": {
+    "strategy": {
+      "Fixed": 300
+    },
+    "direction": "Right",
+    "pad_to_multiple_of": null,
+    "pad_id": 50280,
+    "pad_type_id": 0,
+    "pad_token": "[PAD]"
+  },
   "added_tokens": [
     {
       "id": 0,
@@ -293,6 +307,7 @@
     "continuing_subword_prefix": null,
     "end_of_word_suffix": null,
     "fuse_unk": false,
+    "byte_fallback": false,
     "vocab": {
       "<|endoftext|>": 0,
       "<|padding|>": 1,

tokenizer_config.json

@@ -1,10 +1,9 @@
 {
   "add_prefix_space": false,
   "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
   "eos_token": "<|endoftext|>",
-  "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "EleutherAI/pythia-410m",
-  "special_tokens_map_file": "/admin/home-hailey/.cache/huggingface/hub/models--EleutherAI--gpt-neox-20b/snapshots/4e49eadb5d14bd22f314ec3f45b69a87b88c7691/special_tokens_map.json",
+  "model_max_length": 300,
   "tokenizer_class": "GPTNeoXTokenizer",
   "unk_token": "<|endoftext|>"
 }