fix: update pad token in tokenizer following upstream
Browse files- tokenizer_config.json +1 -9
tokenizer_config.json
CHANGED
@@ -9071,14 +9071,6 @@
|
|
9071 |
"rstrip": false,
|
9072 |
"single_word": false,
|
9073 |
"special": true
|
9074 |
-
},
|
9075 |
-
"201134": {
|
9076 |
-
"content": "<|finetune_right_pad_id|>",
|
9077 |
-
"lstrip": false,
|
9078 |
-
"normalized": false,
|
9079 |
-
"rstrip": false,
|
9080 |
-
"single_word": false,
|
9081 |
-
"special": true
|
9082 |
}
|
9083 |
},
|
9084 |
"bos_token": "<|begin_of_text|>",
|
@@ -9090,7 +9082,7 @@
|
|
9090 |
"attention_mask"
|
9091 |
],
|
9092 |
"model_max_length": 262144,
|
9093 |
-
"pad_token": "<|
|
9094 |
"processor_class": "Llama4Processor",
|
9095 |
"tokenizer_class": "PreTrainedTokenizer"
|
9096 |
}
|
|
|
9071 |
"rstrip": false,
|
9072 |
"single_word": false,
|
9073 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9074 |
}
|
9075 |
},
|
9076 |
"bos_token": "<|begin_of_text|>",
|
|
|
9082 |
"attention_mask"
|
9083 |
],
|
9084 |
"model_max_length": 262144,
|
9085 |
+
"pad_token": "<|finetune_right_pad|>",
|
9086 |
"processor_class": "Llama4Processor",
|
9087 |
"tokenizer_class": "PreTrainedTokenizer"
|
9088 |
}
|