Upload folder using huggingface_hub

added_tokens.json

@@ -2,7 +2,7 @@
   "</s>": 2,
   "<s>": 1,
   "<unk>": 0,
-  "<|bof|>": 32000,
+  "<|bof|>": 32002,
   "<|eof|>": 32001,
-  "<|eot|>": 32002
+  "<|eot|>": 32000
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "mistralai/Mistral-7B-v0.1",
+  "_name_or_path": "/workspace/Mistral-7B-v0.1",
   "architectures": [
     "MistralForCausalLM"
   ],
@@ -14,7 +14,6 @@
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
   "num_key_value_heads": 8,
-  "pad_token_id": 2,
   "rms_norm_eps": 1e-05,
   "rope_theta": 10000.0,
   "sliding_window": 4096,

pytorch_model-00001-of-00002.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f23e8d1e200efb195517a3ef9cb3efea3f8276ead85a82a860abd792c60e1096
+oid sha256:552ad84e3a3ec166a1ab4ab5166276ed4ba95b9c9592e5ff46c85ef7a68a0b3c
 size 9943055436

pytorch_model-00002-of-00002.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d63a15594b099674cacfcbab53949209bbe973220568c9eaa5dd842e782e94a
+oid sha256:2fe74e959b70e2732c2ed276630de2e091473ace6c47f80311e1fdd6fbb1f54a
 size 4540561439

special_tokens_map.json

@@ -2,10 +2,12 @@
   "additional_special_tokens": [
     "<unk>",
     "<s>",
-    "</s>"
+    "</s>",
+    "<|eot|>",
+    "<|eof|>",
+    "<|bof|>"
   ],
   "bos_token": "<s>",
   "eos_token": "</s>",
-  "pad_token": "</s>",
   "unk_token": "<unk>"
 }

tokenizer_config.json

@@ -27,45 +27,48 @@
       "special": true
     },
     "32000": {
-      "content": "<|bof|>",
+      "content": "<|eot|>",
       "lstrip": true,
-      "normalized": true,
+      "normalized": false,
       "rstrip": true,
       "single_word": false,
-      "special": false
+      "special": true
     },
     "32001": {
       "content": "<|eof|>",
       "lstrip": true,
-      "normalized": true,
+      "normalized": false,
       "rstrip": true,
       "single_word": false,
-      "special": false
+      "special": true
     },
     "32002": {
-      "content": "<|eot|>",
+      "content": "<|bof|>",
       "lstrip": true,
-      "normalized": true,
+      "normalized": false,
       "rstrip": true,
       "single_word": false,
-      "special": false
+      "special": true
     }
   },
   "additional_special_tokens": [
     "<unk>",
     "<s>",
-    "</s>"
+    "</s>",
+    "<|eot|>",
+    "<|eof|>",
+    "<|bof|>"
   ],
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "legacy": true,
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "</s>",
+  "pad_token": null,
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "LlamaTokenizer",
-  "tokenizer_file": "/root/.cache/huggingface/hub/models--mistralai--Mistral-7B-v0.1/snapshots/60e27e7bebd4aaf2edf6f4d172346bfc95e2ae40/tokenizer.json",
+  "tokenizer_file": "/workspace/Mistral-7B-v0.1/tokenizer.json",
   "unk_token": "<unk>",
   "use_default_system_prompt": true
 }