diff --git a/.gitattributes b/.gitattributes
index fab8cb33296b0ebe0ed5e6bc8b817b83dad325e5..f07fa1247ac240f8e097523d96fc8a47423ebd0c 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -50,3 +50,12 @@ checkpoints/za_CN-zh_CN/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/za_CN-zh_CN-switchout/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/zh_CN-za_CN-switchout/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/za-zh-synonyms/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/za-zh-theme/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/zh-za-theme/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/zh-za-synonyms/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/za-zh-rs-za/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/za-zh-rd-za/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/za-zh-ri-za/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/za_zh_sc/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/zh_za_sc/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/za-zh-tssr-za/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/za-zh-rd-za/.gitattributes b/checkpoints/za-zh-rd-za/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/za-zh-rd-za/README.md b/checkpoints/za-zh-rd-za/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..faa0d94b54b401d85e9a6a0293abf8c4f4382f93
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/README.md
@@ -0,0 +1,63 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: za-zh-rd-za
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# za-zh-rd-za
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.9474
+- Sacrebleu: 4.4145
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 1.5105        | 1.0   | 1185 | 2.8866          | 2.3746    |
+| 0.7597        | 2.0   | 2370 | 2.8919          | 2.4519    |
+| 0.4917        | 3.0   | 3555 | 2.9474          | 4.4145    |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/za-zh-rd-za/config.json b/checkpoints/za-zh-rd-za/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..22df9f6481eb128d1cd587a740d206dd18c57bfe
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250054
+}
diff --git a/checkpoints/za-zh-rd-za/generation_config.json b/checkpoints/za-zh-rd-za/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/za-zh-rd-za/model.safetensors b/checkpoints/za-zh-rd-za/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..33af4c3e2ba99fee2253a9cdae4bd1c7e95d0c3d
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77ac7f2f1fffd22e6ddf23bddbfc9bb2f345f6825ac64ce966bb9b96e6d216a7
+size 2444578688
diff --git a/checkpoints/za-zh-rd-za/tokenizer.json b/checkpoints/za-zh-rd-za/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21e0e513bfad7883422c75ca167b7608de858402
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e673ccd228f195203d0ca7520a3e75b660426c6d0cda003d28f43d49069a48d8
+size 16793291
diff --git a/checkpoints/za-zh-rd-za/tokenizer_config.json b/checkpoints/za-zh-rd-za/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f00a7aee0c9a2c816cc8bc77915b89dc449c99d2
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/tokenizer_config.json
@@ -0,0 +1,73 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<s>",
+    "</s>",
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/za-zh-rd-za/training_args.bin b/checkpoints/za-zh-rd-za/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..93ad9e8129da59404c237b7c23c9a74d99652f25
--- /dev/null
+++ b/checkpoints/za-zh-rd-za/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d273db7d95286e56942812d0cf9eab28029a22dde0b56a1c4306a6a3d9cd8ee
+size 5393
diff --git a/checkpoints/za-zh-ri-za/.gitattributes b/checkpoints/za-zh-ri-za/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/za-zh-ri-za/README.md b/checkpoints/za-zh-ri-za/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..6de96a039732e5218e7fb3bdbc27794030bbc1c2
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/README.md
@@ -0,0 +1,63 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: za-zh-ri-za
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# za-zh-ri-za
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.8529
+- Sacrebleu: 4.6252
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 0.8773        | 1.0   | 1545 | 2.7437          | 3.8543    |
+| 0.2678        | 2.0   | 3090 | 2.8529          | 4.6252    |
+| 0.1240        | 3.0   | 4635 | 2.9385          | 3.6222    |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/za-zh-ri-za/config.json b/checkpoints/za-zh-ri-za/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..22df9f6481eb128d1cd587a740d206dd18c57bfe
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250054
+}
diff --git a/checkpoints/za-zh-ri-za/generation_config.json b/checkpoints/za-zh-ri-za/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/za-zh-ri-za/model.safetensors b/checkpoints/za-zh-ri-za/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..102e65dae2beb62891395d3710a4123aa2739ce1
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7be6dd690a158b9febc5c3b73258d952faca884743456e30fac3c7bafbc55d62
+size 2444578688
diff --git a/checkpoints/za-zh-ri-za/tokenizer.json b/checkpoints/za-zh-ri-za/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21e0e513bfad7883422c75ca167b7608de858402
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e673ccd228f195203d0ca7520a3e75b660426c6d0cda003d28f43d49069a48d8
+size 16793291
diff --git a/checkpoints/za-zh-ri-za/tokenizer_config.json b/checkpoints/za-zh-ri-za/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f00a7aee0c9a2c816cc8bc77915b89dc449c99d2
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/tokenizer_config.json
@@ -0,0 +1,73 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<s>",
+    "</s>",
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/za-zh-ri-za/training_args.bin b/checkpoints/za-zh-ri-za/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..126d035f7e8d256ee71198c48dcabb8451fae7d7
--- /dev/null
+++ b/checkpoints/za-zh-ri-za/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0101d6469f336a9c3a1133f0e59e983735bf2883b8a46cfc6c9993e89462e8e6
+size 5393
diff --git a/checkpoints/za-zh-rs-za/.gitattributes b/checkpoints/za-zh-rs-za/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/za-zh-rs-za/README.md b/checkpoints/za-zh-rs-za/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..883cac774468f1bcafae29c4e74aa4e441874750
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/README.md
@@ -0,0 +1,63 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: za-zh-rs-za
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# za-zh-rs-za
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.2987
+- Sacrebleu: 2.6285
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 1.9199        | 1.0   | 1236 | 3.2717          | 1.9790    |
+| 1.1039        | 2.0   | 2472 | 3.2611          | 0.9655    |
+| 0.7351        | 3.0   | 3708 | 3.2987          | 2.6285    |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/za-zh-rs-za/config.json b/checkpoints/za-zh-rs-za/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..22df9f6481eb128d1cd587a740d206dd18c57bfe
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250054
+}
diff --git a/checkpoints/za-zh-rs-za/generation_config.json b/checkpoints/za-zh-rs-za/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/za-zh-rs-za/model.safetensors b/checkpoints/za-zh-rs-za/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..23053d92216f250fb69218724ce7cf61058a938c
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd05e336e86c84789be5d9ca3b7d527841d5d7cb8db0ebd444ec5459a6793bc8
+size 2444578688
diff --git a/checkpoints/za-zh-rs-za/tokenizer.json b/checkpoints/za-zh-rs-za/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21e0e513bfad7883422c75ca167b7608de858402
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e673ccd228f195203d0ca7520a3e75b660426c6d0cda003d28f43d49069a48d8
+size 16793291
diff --git a/checkpoints/za-zh-rs-za/tokenizer_config.json b/checkpoints/za-zh-rs-za/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f00a7aee0c9a2c816cc8bc77915b89dc449c99d2
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/tokenizer_config.json
@@ -0,0 +1,73 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<s>",
+    "</s>",
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/za-zh-rs-za/training_args.bin b/checkpoints/za-zh-rs-za/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..93810ab4cf8f2507195decd88233b5d23a35738c
--- /dev/null
+++ b/checkpoints/za-zh-rs-za/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:548cff87467b53c387d590852d47224d6903ef95c6b3792f71aa9a43a291a4ac
+size 5393
diff --git a/checkpoints/za-zh-theme/.gitattributes b/checkpoints/za-zh-theme/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/za-zh-theme/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/za-zh-theme/README.md b/checkpoints/za-zh-theme/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..fcaced2fd82d75b1aecbd6297c42927e0de13418
--- /dev/null
+++ b/checkpoints/za-zh-theme/README.md
@@ -0,0 +1,63 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: za-zh-theme
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# za-zh-theme
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.7864
+- Sacrebleu: 6.6442
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 0.4156        | 1.0   | 1678 | 2.7864          | 6.6442    |
+| 0.1474        | 2.0   | 3356 | 2.8686          | 5.0403    |
+| 0.0730        | 3.0   | 5034 | 2.9130          | 6.2466    |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/za-zh-theme/config.json b/checkpoints/za-zh-theme/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..22df9f6481eb128d1cd587a740d206dd18c57bfe
--- /dev/null
+++ b/checkpoints/za-zh-theme/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250054
+}
diff --git a/checkpoints/za-zh-theme/generation_config.json b/checkpoints/za-zh-theme/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/za-zh-theme/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/za-zh-theme/model.safetensors b/checkpoints/za-zh-theme/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..af91ec61df8c66698a8b3e0dd6a3de4b4c446ae7
--- /dev/null
+++ b/checkpoints/za-zh-theme/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ace579aefd498d0ea9ab41882b3d8faae2eaafcf5a1f4b294898bff25c25a5a
+size 2444578688
diff --git a/checkpoints/za-zh-theme/tokenizer.json b/checkpoints/za-zh-theme/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21e0e513bfad7883422c75ca167b7608de858402
--- /dev/null
+++ b/checkpoints/za-zh-theme/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e673ccd228f195203d0ca7520a3e75b660426c6d0cda003d28f43d49069a48d8
+size 16793291
diff --git a/checkpoints/za-zh-theme/tokenizer_config.json b/checkpoints/za-zh-theme/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f00a7aee0c9a2c816cc8bc77915b89dc449c99d2
--- /dev/null
+++ b/checkpoints/za-zh-theme/tokenizer_config.json
@@ -0,0 +1,73 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<s>",
+    "</s>",
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/za-zh-theme/training_args.bin b/checkpoints/za-zh-theme/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..7b98bb2921bcdb3071ef2261a647b773a90e2fb2
--- /dev/null
+++ b/checkpoints/za-zh-theme/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed56aa4a87ea643d4bf09369d4f664069629d656b05e2efe1e8600d91267cde7
+size 5393
diff --git a/checkpoints/za-zh-tssr-za/.gitattributes b/checkpoints/za-zh-tssr-za/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/za-zh-tssr-za/README.md b/checkpoints/za-zh-tssr-za/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..327922db20153f45b6661bddf472ca0fa1c398f1
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/README.md
@@ -0,0 +1,63 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: za-zh-tssr-za
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# za-zh-tssr-za
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4848
+- Sacrebleu: 5.5321
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 0.3205        | 1.0   | 407  | 0.5258          | 5.5128    |
+| 0.1870        | 2.0   | 814  | 0.4896          | 4.1458    |
+| 0.1451        | 3.0   | 1221 | 0.4848          | 5.5321    |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/za-zh-tssr-za/config.json b/checkpoints/za-zh-tssr-za/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..22df9f6481eb128d1cd587a740d206dd18c57bfe
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250054
+}
diff --git a/checkpoints/za-zh-tssr-za/generation_config.json b/checkpoints/za-zh-tssr-za/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/za-zh-tssr-za/model.safetensors b/checkpoints/za-zh-tssr-za/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fc765470f223aa0c9a0dbc0dbf69be7e0dfa809f
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:526c866f7219de54f8805eb6a74a36554feb25cfac41bfd4e5a1026da39d4c6c
+size 2444578688
diff --git a/checkpoints/za-zh-tssr-za/tokenizer.json b/checkpoints/za-zh-tssr-za/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21e0e513bfad7883422c75ca167b7608de858402
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e673ccd228f195203d0ca7520a3e75b660426c6d0cda003d28f43d49069a48d8
+size 16793291
diff --git a/checkpoints/za-zh-tssr-za/tokenizer_config.json b/checkpoints/za-zh-tssr-za/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f00a7aee0c9a2c816cc8bc77915b89dc449c99d2
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/tokenizer_config.json
@@ -0,0 +1,73 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<s>",
+    "</s>",
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/za-zh-tssr-za/training_args.bin b/checkpoints/za-zh-tssr-za/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a1327146a41df613887fa7952151bc10eab79f5f
--- /dev/null
+++ b/checkpoints/za-zh-tssr-za/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2736c870aeb7071be9641339d024aff43187bd624ff5e779a1dcb0058fe62556
+size 5393
diff --git a/checkpoints/za_zh_sc/.gitattributes b/checkpoints/za_zh_sc/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/za_zh_sc/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/za_zh_sc/README.md b/checkpoints/za_zh_sc/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..2e006adf7471e5556b60e509c6dfa11e63ab4982
--- /dev/null
+++ b/checkpoints/za_zh_sc/README.md
@@ -0,0 +1,62 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: za_zh_sc
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# za_zh_sc
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.9950
+- Sacrebleu: 4.4405
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 2
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 2.5108        | 1.0   | 309  | 2.9950          | 4.4405    |
+| 1.7678        | 2.0   | 618  | 2.8134          | 3.4752    |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/za_zh_sc/config.json b/checkpoints/za_zh_sc/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..56794146c368c92a0d7720053e5e9d8bbe90d9c4
--- /dev/null
+++ b/checkpoints/za_zh_sc/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250055
+}
diff --git a/checkpoints/za_zh_sc/generation_config.json b/checkpoints/za_zh_sc/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/za_zh_sc/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/za_zh_sc/model.safetensors b/checkpoints/za_zh_sc/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2e90838a7702be502568a16592e119d51ac7b826
--- /dev/null
+++ b/checkpoints/za_zh_sc/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0e7cc73ae9b3435c2455575f9d3217bdf54f799f014084d890f3f42b524e902
+size 2444582788
diff --git a/checkpoints/za_zh_sc/tokenizer.json b/checkpoints/za_zh_sc/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..c76a74afe27fad56f908b9b90a6c04530812245b
--- /dev/null
+++ b/checkpoints/za_zh_sc/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1eeb5a28135f15f64314077af5cae74547d898c979266144ecd95f709e76b008
+size 16793473
diff --git a/checkpoints/za_zh_sc/tokenizer_config.json b/checkpoints/za_zh_sc/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e5e060af7ddd3bf07ea5d94b414e38cca441be47
--- /dev/null
+++ b/checkpoints/za_zh_sc/tokenizer_config.json
@@ -0,0 +1,20 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<SEP>"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/za_zh_sc/training_args.bin b/checkpoints/za_zh_sc/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..0abfe948b635699dbd9a17c693986c8edf6b2f90
--- /dev/null
+++ b/checkpoints/za_zh_sc/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c86211a54299f4647a5833604b3acbcad70a287337769adf2678c85bc368b635
+size 5329
diff --git a/checkpoints/zh-za-synonyms/.gitattributes b/checkpoints/zh-za-synonyms/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/zh-za-synonyms/README.md b/checkpoints/zh-za-synonyms/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..902d176c5965f75b256582d0f95e034b42dfcaae
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/README.md
@@ -0,0 +1,63 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: zh-za-synonyms
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# zh-za-synonyms
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.3693
+- Sacrebleu: 18.8842
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 0.4186        | 1.0   | 2028 | 2.2081          | 16.6134   |
+| 0.1665        | 2.0   | 4056 | 2.3110          | 17.7182   |
+| 0.0988        | 3.0   | 6084 | 2.3693          | 18.8842   |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/zh-za-synonyms/config.json b/checkpoints/zh-za-synonyms/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..22df9f6481eb128d1cd587a740d206dd18c57bfe
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250054
+}
diff --git a/checkpoints/zh-za-synonyms/generation_config.json b/checkpoints/zh-za-synonyms/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/zh-za-synonyms/model.safetensors b/checkpoints/zh-za-synonyms/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..27105b0c227cd1ea4b41d3aca359739178511802
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f83c750baf685bfa98641d31b63085f1cc00ee4b15ff861923d8b17aa022d809
+size 2444578688
diff --git a/checkpoints/zh-za-synonyms/tokenizer.json b/checkpoints/zh-za-synonyms/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21e0e513bfad7883422c75ca167b7608de858402
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e673ccd228f195203d0ca7520a3e75b660426c6d0cda003d28f43d49069a48d8
+size 16793291
diff --git a/checkpoints/zh-za-synonyms/tokenizer_config.json b/checkpoints/zh-za-synonyms/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f00a7aee0c9a2c816cc8bc77915b89dc449c99d2
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/tokenizer_config.json
@@ -0,0 +1,73 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<s>",
+    "</s>",
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/zh-za-synonyms/training_args.bin b/checkpoints/zh-za-synonyms/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..ba9dbd14b047cd8b2ec409b5e2bdbd8cc894f9c1
--- /dev/null
+++ b/checkpoints/zh-za-synonyms/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69fed2656a391fe17c3f4098e355a568d38acb9e3b0bb4591d2a00decdd63b7a
+size 5393
diff --git a/checkpoints/zh-za-theme/.gitattributes b/checkpoints/zh-za-theme/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/zh-za-theme/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/zh-za-theme/README.md b/checkpoints/zh-za-theme/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..018e3f90e4365837e792ef2b3fddd0207caabcad
--- /dev/null
+++ b/checkpoints/zh-za-theme/README.md
@@ -0,0 +1,63 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: zh-za-theme
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# zh-za-theme
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.0812
+- Sacrebleu: 19.5320
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 0.7090        | 1.0   | 1678 | 2.0409          | 16.4536   |
+| 0.4013        | 2.0   | 3356 | 2.0482          | 18.7315   |
+| 0.2852        | 3.0   | 5034 | 2.0812          | 19.5320   |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/zh-za-theme/config.json b/checkpoints/zh-za-theme/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..22df9f6481eb128d1cd587a740d206dd18c57bfe
--- /dev/null
+++ b/checkpoints/zh-za-theme/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250054
+}
diff --git a/checkpoints/zh-za-theme/generation_config.json b/checkpoints/zh-za-theme/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/zh-za-theme/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/zh-za-theme/model.safetensors b/checkpoints/zh-za-theme/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..169031560a7b69c37047a3753f168a28c2d79209
--- /dev/null
+++ b/checkpoints/zh-za-theme/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e25e745178f07e30652538f9e6db6f57c6d99b11a1c01c43441a55f399a6fe8
+size 2444578688
diff --git a/checkpoints/zh-za-theme/tokenizer.json b/checkpoints/zh-za-theme/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..21e0e513bfad7883422c75ca167b7608de858402
--- /dev/null
+++ b/checkpoints/zh-za-theme/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e673ccd228f195203d0ca7520a3e75b660426c6d0cda003d28f43d49069a48d8
+size 16793291
diff --git a/checkpoints/zh-za-theme/tokenizer_config.json b/checkpoints/zh-za-theme/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f00a7aee0c9a2c816cc8bc77915b89dc449c99d2
--- /dev/null
+++ b/checkpoints/zh-za-theme/tokenizer_config.json
@@ -0,0 +1,73 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<s>",
+    "</s>",
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/zh-za-theme/training_args.bin b/checkpoints/zh-za-theme/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d8c8237f8e1c19459181d219e8b5849a8f30e9d4
--- /dev/null
+++ b/checkpoints/zh-za-theme/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2b74bc0376fb976e4d867d958c95c98c2b47ac5b9c05efd826a3220ad2305c0
+size 5393
diff --git a/checkpoints/zh_za_sc/.gitattributes b/checkpoints/zh_za_sc/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..52373fe24473b1aa44333d318f578ae6bf04b49b
--- /dev/null
+++ b/checkpoints/zh_za_sc/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/checkpoints/zh_za_sc/README.md b/checkpoints/zh_za_sc/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..7a74851e53348220245a852f1f234294ac0fc54f
--- /dev/null
+++ b/checkpoints/zh_za_sc/README.md
@@ -0,0 +1,62 @@
+---
+library_name: transformers
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- sacrebleu
+model-index:
+- name: zh_za_sc
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# zh_za_sc
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.4201
+- Sacrebleu: 7.7132
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 2
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|
+| 2.4518        | 1.0   | 309  | 2.6494          | 4.2779    |
+| 1.8813        | 2.0   | 618  | 2.4201          | 7.7132    |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2
diff --git a/checkpoints/zh_za_sc/config.json b/checkpoints/zh_za_sc/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..56794146c368c92a0d7720053e5e9d8bbe90d9c4
--- /dev/null
+++ b/checkpoints/zh_za_sc/config.json
@@ -0,0 +1,55 @@
+{
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_decoder": false,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tie_word_embeddings": true,
+  "tokenizer_class": "MBart50Tokenizer",
+  "transformers_version": "5.0.0",
+  "use_cache": false,
+  "vocab_size": 250055
+}
diff --git a/checkpoints/zh_za_sc/generation_config.json b/checkpoints/zh_za_sc/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..0b73e73f7dc705246534a206b75dbd415e4dcc15
--- /dev/null
+++ b/checkpoints/zh_za_sc/generation_config.json
@@ -0,0 +1,37 @@
+{
+  "_from_model_config": false,
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": true,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
+  "eos_token_id": 2,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_eos_token_id": 2,
+  "length_penalty": 1.0,
+  "max_length": 200,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 5,
+  "num_return_sequences": 1,
+  "output_scores": false,
+  "pad_token_id": 1,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
+  "target_lookbehind": 10,
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
+}
diff --git a/checkpoints/zh_za_sc/model.safetensors b/checkpoints/zh_za_sc/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9008761af41c05aae43f054c8e458ede9e5e4209
--- /dev/null
+++ b/checkpoints/zh_za_sc/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c06a5e1baf046fff2c3af2897fca4b018eae0459eade79905fa644b4a384c9bc
+size 2444582788
diff --git a/checkpoints/zh_za_sc/tokenizer.json b/checkpoints/zh_za_sc/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..c76a74afe27fad56f908b9b90a6c04530812245b
--- /dev/null
+++ b/checkpoints/zh_za_sc/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1eeb5a28135f15f64314077af5cae74547d898c979266144ecd95f709e76b008
+size 16793473
diff --git a/checkpoints/zh_za_sc/tokenizer_config.json b/checkpoints/zh_za_sc/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e5e060af7ddd3bf07ea5d94b414e38cca441be47
--- /dev/null
+++ b/checkpoints/zh_za_sc/tokenizer_config.json
@@ -0,0 +1,20 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": [
+    "<SEP>"
+  ],
+  "is_local": false,
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "src_lang": "zh_CN",
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_id": 0,
+  "unk_token": "<unk>"
+}
diff --git a/checkpoints/zh_za_sc/training_args.bin b/checkpoints/zh_za_sc/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..a9683bba5a527580df8d691ce0e32701a3347b31
--- /dev/null
+++ b/checkpoints/zh_za_sc/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:703c39dac1a70a34bb447f2c5853c2c31187d7ff05ba7d88f0e938636e75183b
+size 5329