Training in progress, step 500

config.json

@@ -1,5 +1,6 @@
 {
-  "_name_or_path": "Helsinki-NLP/opus-mt-en-ar",
+  "_name_or_path": "Helsinki-NLP/opus-mt-ar-en",
+  "_num_labels": 3,
   "activation_dropout": 0.0,
   "activation_function": "swish",
   "add_bias_logits": false,
@@ -10,7 +11,7 @@
   "attention_dropout": 0.0,
   "bad_words_ids": [
     [
-      62801
+      62833
     ]
   ],
   "bos_token_id": 0,
@@ -21,15 +22,14 @@
   "decoder_ffn_dim": 2048,
   "decoder_layerdrop": 0.0,
   "decoder_layers": 6,
-  "decoder_start_token_id": 62801,
-  "decoder_vocab_size": 62802,
+  "decoder_start_token_id": 62833,
+  "decoder_vocab_size": 62834,
   "dropout": 0.1,
   "encoder_attention_heads": 8,
   "encoder_ffn_dim": 2048,
   "encoder_layerdrop": 0.0,
   "encoder_layers": 6,
   "eos_token_id": 0,
-  "extra_pos_embeddings": 62802,
   "forced_eos_token_id": 0,
   "id2label": {
     "0": "LABEL_0",
@@ -50,12 +50,12 @@
   "normalize_embedding": false,
   "num_beams": 4,
   "num_hidden_layers": 6,
-  "pad_token_id": 62801,
+  "pad_token_id": 62833,
   "scale_embedding": true,
   "share_encoder_decoder_embeddings": true,
   "static_position_embeddings": true,
   "torch_dtype": "float32",
   "transformers_version": "4.40.0",
   "use_cache": true,
-  "vocab_size": 62802
+  "vocab_size": 62834
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4a4387a3ce8e65e6347ca82507fbaf6f95d8590e04a2d4b38d3c4eb15501a27
-size 305452744
+oid sha256:e8eabadbb3ba1aedc1f5a3d37c0a7a306c55530b3f6a9580036d7c8ef954ef66
+size 305518408

runs/Apr29_09-31-41_876a280947ef/events.out.tfevents.1714383102.876a280947ef.6083.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3cd7f9405b4a8cfdb520a5b75f1c69af036a50c9b7ccfce08a355231224b02d
+size 5800

tokenizer_config.json

@@ -16,7 +16,7 @@
       "single_word": false,
       "special": true
     },
-    "62801": {
+    "62833": {
       "content": "<pad>",
       "lstrip": false,
       "normalized": false,
@@ -30,9 +30,9 @@
   "model_max_length": 512,
   "pad_token": "<pad>",
   "separate_vocabs": false,
-  "source_lang": "eng",
+  "source_lang": "ar",
   "sp_model_kwargs": {},
-  "target_lang": "ara",
+  "target_lang": "en",
   "tokenizer_class": "MarianTokenizer",
   "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0175c2cc4e5f4dacf9bd2af57d4819d8d0e58b80ace441caffc727b63620c2f1
+oid sha256:0eff84af673c7a8aad1db01cf940b5efe299e4169d3aaa61ca1442f3edcd3ebd
 size 5176