Upload version with tags in posts shuffled before training

This model was trained on the same dataset but the tags from each post were shuffled before used for training

config.json

@@ -40,7 +40,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.27.0.dev0",
+  "transformers_version": "4.31.0.dev0",
   "use_cache": true,
   "vocab_size": 50257
 }

merges.txt

@@ -1,4 +1,4 @@
-#version: 0.2 - Trained by `huggingface/tokenizers`
+#version: 0.2
 Ġ t
 Ġ a
 h e

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6353ddc64072be10bc35861e3cc73e639debd29ddfd763ce379bd894004e84de
-size 333970169
+oid sha256:99e39f00ff01123155098e7ec9fdc36432ba403c5d4182a59f2affb76d05a828
+size 327674773

runs/Jun08_11-12-20_671aa6aed209/events.out.tfevents.1686222803.671aa6aed209.1153.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a2d97deabcf9a45ed754f0fcd585686c3b89b48c616d33e9205d00d3dfe9284
+size 6931

runs/Jun08_11-12-20_671aa6aed209/events.out.tfevents.1686232154.671aa6aed209.1153.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1a15fd04acb34741f690785349d2fdaa82f9627a84c18ff520b73509541313a
+size 411

tokenizer.json

@@ -39,6 +39,7 @@
     "continuing_subword_prefix": "",
     "end_of_word_suffix": "",
     "fuse_unk": false,
+    "byte_fallback": false,
     "vocab": {
       "!": 0,
       "\"": 1,

tokenizer_config.json

@@ -1,9 +1,9 @@
 {
   "add_prefix_space": false,
   "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
   "eos_token": "<|endoftext|>",
   "model_max_length": 1024,
-  "special_tokens_map_file": null,
   "tokenizer_class": "GPT2Tokenizer",
   "unk_token": "<|endoftext|>"
 }