Canary-1b-v2 tokenizer backwards compitability
Browse files
canary-1b-v2/config.json
CHANGED
|
@@ -29,9 +29,10 @@
|
|
| 29 |
"subsampling_conv_stride": 2,
|
| 30 |
"subsampling_factor": 8
|
| 31 |
},
|
|
|
|
|
|
|
|
|
|
| 32 |
"initializer_range": 0.02,
|
| 33 |
-
"model_type": "parakeet_ctc",
|
| 34 |
"pad_token_id": 1024,
|
| 35 |
-
"transformers_version": "4.57.1",
|
| 36 |
"vocab_size": 16385
|
| 37 |
}
|
|
|
|
| 29 |
"subsampling_conv_stride": 2,
|
| 30 |
"subsampling_factor": 8
|
| 31 |
},
|
| 32 |
+
"bos_token_id": 4,
|
| 33 |
+
"eos_token_id": 3,
|
| 34 |
+
"nemo_model_type": "parakeet",
|
| 35 |
"initializer_range": 0.02,
|
|
|
|
| 36 |
"pad_token_id": 1024,
|
|
|
|
| 37 |
"vocab_size": 16385
|
| 38 |
}
|
canary-1b-v2/tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
canary-1b-v2/tokenizer_config.json
CHANGED
|
@@ -9310,6 +9310,6 @@
|
|
| 9310 |
"model_max_length": 1000000000000000019884624838656,
|
| 9311 |
"pad_token": "<pad>",
|
| 9312 |
"processor_class": "ParakeetProcessor",
|
| 9313 |
-
"tokenizer_class": "
|
| 9314 |
"unk_token": "<unk>"
|
| 9315 |
}
|
|
|
|
| 9310 |
"model_max_length": 1000000000000000019884624838656,
|
| 9311 |
"pad_token": "<pad>",
|
| 9312 |
"processor_class": "ParakeetProcessor",
|
| 9313 |
+
"tokenizer_class": "PreTrainedTokenizer",
|
| 9314 |
"unk_token": "<unk>"
|
| 9315 |
}
|