Training in progress, step 500

added_tokens.json

@@ -1,4 +1,4 @@
 {
-  "</s>": 73,
-  "<s>": 72
+  "</s>": 76,
+  "<s>": 75
 }

config.json

@@ -9,7 +9,7 @@
   "architectures": [
     "Wav2Vec2BertForCTC"
   ],
-  "attention_dropout": 0.0,
+  "attention_dropout": 0.1,
   "blank_token_id": 0,
   "bos_token_id": 1,
   "classifier_proj_size": 768,
@@ -26,7 +26,7 @@
   "feature_projection_input_dim": 160,
   "final_dropout": 0.1,
   "hidden_act": "swish",
-  "hidden_dropout": 0.0,
+  "hidden_dropout": 0.1,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
@@ -48,7 +48,7 @@
   "num_hidden_layers": 24,
   "num_negatives": 100,
   "output_hidden_size": 1024,
-  "pad_token_id": 68,
+  "pad_token_id": 71,
   "position_embeddings_type": "relative_key",
   "proj_codevector_dim": 768,
   "right_max_position_embeddings": 8,
@@ -78,6 +78,6 @@
   "transformers_version": "4.49.0",
   "use_intermediate_ffn_before_adapter": false,
   "use_weighted_layer_sum": false,
-  "vocab_size": 74,
+  "vocab_size": 77,
   "xvector_output_dim": 512
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de6262a0e09af1c4f8de3b03b21577cf8b37ab964d5951862b84fa2cc52c9f42
-size 2322378112
+oid sha256:f861f5ef22b98759dbd153a1f06f9dc819cce350e445c0354823b68fec3a968f
+size 2322390412

tokenizer_config.json

@@ -24,7 +24,7 @@
       "single_word": false,
       "special": false
     },
-    "68": {
+    "71": {
       "content": "[PAD]",
       "lstrip": true,
       "normalized": false,
@@ -32,7 +32,7 @@
       "single_word": false,
       "special": false
     },
-    "69": {
+    "72": {
       "content": "[UNK]",
       "lstrip": true,
       "normalized": false,
@@ -40,7 +40,7 @@
       "single_word": false,
       "special": false
     },
-    "72": {
+    "75": {
       "content": "<s>",
       "lstrip": false,
       "normalized": false,
@@ -48,7 +48,7 @@
       "single_word": false,
       "special": true
     },
-    "73": {
+    "76": {
       "content": "</s>",
       "lstrip": false,
       "normalized": false,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f83e6c3efd60d97448ee0ad04f217955bbc10e33fc5a49005ee4cd09e9fb81a
+oid sha256:8e8228845e62a44b76c6d252121dd5b91007f4f43129c6a5cc9b250bacfa2e91
 size 5368

vocab.json

@@ -1,74 +1,77 @@
 {
-  "(": 60,
-  ")": 61,
-  ",": 20,
+  "!": 67,
+  ",": 19,
   "-": 38,
   ".": 26,
-  "7": 57,
-  ":": 36,
-  ";": 62,
-  "<": 34,
+  ":": 33,
+  "<": 36,
   "<ctc_blank>": 0,
   "<pad>": 1,
   "<unk>": 2,
-  ">": 35,
-  "?": 49,
-  "[PAD]": 68,
-  "[UNK]": 69,
-  "А": 27,
+  ">": 37,
+  "?": 53,
+  "[PAD]": 71,
+  "[UNK]": 72,
+  "«": 54,
+  "»": 55,
+  "А": 29,
   "Б": 45,
+  "В": 63,
   "Г": 39,
-  "Д": 48,
-  "Е": 55,
-  "И": 63,
-  "К": 41,
-  "Л": 50,
+  "Д": 44,
+  "Е": 58,
+  "Ж": 68,
+  "И": 56,
+  "К": 40,
+  "Л": 59,
   "М": 46,
-  "Н": 58,
-  "О": 53,
-  "П": 64,
-  "Р": 44,
-  "С": 52,
+  "Н": 49,
+  "О": 57,
+  "П": 69,
+  "Р": 43,
+  "С": 50,
   "Т": 47,
-  "У": 59,
-  "Х": 43,
-  "Ц": 65,
-  "Э": 66,
-  "Ю": 67,
+  "У": 64,
+  "Ф": 65,
+  "Х": 51,
+  "Ц": 66,
+  "Э": 62,
+  "Ю": 70,
   "а": 4,
   "б": 17,
-  "в": 25,
+  "в": 24,
   "г": 11,
   "д": 23,
-  "е": 8,
+  "е": 7,
   "ж": 28,
   "з": 21,
   "и": 9,
-  "й": 24,
+  "й": 25,
   "к": 14,
-  "л": 7,
-  "м": 22,
+  "л": 8,
+  "м": 20,
   "н": 10,
   "о": 5,
-  "п": 37,
+  "п": 42,
   "р": 13,
   "с": 18,
-  "т": 29,
-  "у": 16,
-  "ф": 56,
-  "х": 15,
+  "т": 27,
+  "у": 15,
+  "ф": 48,
+  "х": 16,
   "ц": 31,
   "ч": 32,
-  "ш": 33,
-  "щ": 51,
+  "ш": 34,
+  "щ": 60,
   "ъ": 6,
   "ь": 12,
   "э": 30,
-  "ю": 54,
-  "я": 40,
-  "ё": 42,
-  "Ӏ": 19,
-  "‌": 70,
-  "‍": 71,
+  "ю": 52,
+  "я": 35,
+  "ё": 41,
+  "Ӏ": 22,
+  "‌": 73,
+  "‍": 74,
+  "–": 61,
   "▁": 3
 }