Saving weights. Last push

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "### Fim": 40559,
+  "### Instrução:": 40560,
+  "### Resposta:\n": 40561
+}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "thegoodfellas/tgf-gpt-117m",
+  "_name_or_path": "thegoodfellas/tgf-gpt-117m-mc4",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
@@ -32,7 +32,7 @@
       "max_length": 50
     }
   },
-  "transformers_version": "4.26.1",
+  "transformers_version": "4.28.1",
   "use_cache": true,
-  "vocab_size": 50257
+  "vocab_size": 50260
 }

flax_model.msgpack

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd04f2494647bb6051d2963f92441ffc7596248931d843254a6aee3a0eb030d7
-size 497764120
+oid sha256:90fb0bebbdef5e8894375dfb92408d7b5b448220ead5a80c4585a9c93a108c8b
+size 497773336

generation_config.json

@@ -2,5 +2,5 @@
   "_from_model_config": true,
   "bos_token_id": 50256,
   "eos_token_id": 50256,
-  "transformers_version": "4.26.1"
+  "transformers_version": "4.28.1"
 }

merges.txt

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

special_tokens_map.json

@@ -1,4 +1,9 @@
 {
+  "additional_special_tokens": [
+    "### Fim",
+    "### Instrução:",
+    "### Resposta:\n"
+  ],
   "bos_token": "<s>",
   "cls_token": "<s>",
   "eos_token": "</s>",
@@ -9,7 +14,7 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<pad>",
+  "pad_token": "</s>",
   "sep_token": "</s>",
   "unk_token": "<unk>"
 }

tokenizer.json

@@ -47,6 +47,33 @@
       "rstrip": false,
       "normalized": false,
       "special": true
+    },
+    {
+      "id": 40559,
+      "content": "### Fim",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 40560,
+      "content": "### Instrução:",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 40561,
+      "content": "### Resposta:\n",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
     }
   ],
   "normalizer": null,
@@ -75,6 +102,7 @@
     "continuing_subword_prefix": null,
     "end_of_word_suffix": null,
     "fuse_unk": false,
+    "byte_fallback": false,
     "vocab": {
       "<s>": 0,
       "<pad>": 1,

tokenizer_config.json

@@ -1,15 +1,14 @@
 {
   "add_prefix_space": false,
   "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
   "cls_token": "<s>",
   "eos_token": "</s>",
   "errors": "replace",
   "mask_token": "<mask>",
   "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "thegoodfellas/tgf-gpt-117m",
   "pad_token": "<pad>",
   "sep_token": "</s>",
-  "special_tokens_map_file": null,
   "tokenizer_class": "RobertaTokenizer",
   "trim_offsets": true,
   "unk_token": "<unk>"