Training in progress, step 500

config.json

@@ -52,7 +52,7 @@
   "static_position_embeddings": false,
   "tokenizer_class": "MBart50Tokenizer",
   "torch_dtype": "float32",
-  "transformers_version": "4.26.1",
+  "transformers_version": "4.33.1",
   "use_cache": true,
   "vocab_size": 250054
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5bb893a330c79801e7362d2e12c84784fd8a9032d481fcaf53e4d083f2d394f
+oid sha256:967cc4559bc8d12d5e18e3e129872e4db6d211b10f54331764cd6a9173c647c7
 size 2444694045

tokenizer_config.json

@@ -54,6 +54,7 @@
     "sl_SI"
   ],
   "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
   "cls_token": "<s>",
   "eos_token": "</s>",
   "mask_token": {
@@ -65,11 +66,9 @@
     "single_word": false
   },
   "model_max_length": 1024,
-  "name_or_path": "facebook/mbart-large-50",
   "pad_token": "<pad>",
   "sep_token": "</s>",
   "sp_model_kwargs": {},
-  "special_tokens_map_file": "/home/suraj/projects/mbart-50/hf_models/mbart-50-large/special_tokens_map.json",
   "src_lang": null,
   "tgt_lang": null,
   "tokenizer_class": "MBart50Tokenizer",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9acc751cc740415ae0fd6c4c18d67a48d5667b241c9e17d6e2cfbcd592519184
-size 3643
+oid sha256:26f09fd50a532cddc84a8db6679364939d8bd19b5312b8373a68f56bd5f503c1
+size 4155