Training in progress, step 1000

added_tokens.json

@@ -17,7 +17,6 @@
   "<|da|>": 50285,
   "<|de|>": 50261,
   "<|el|>": 50281,
-  "<|endoftext|>": 50257,
   "<|en|>": 50259,
   "<|es|>": 50262,
   "<|et|>": 50307,
@@ -30,6 +29,7 @@
   "<|gu|>": 50333,
   "<|haw|>": 50352,
   "<|ha|>": 50354,
+  "<|he|>": 50279,
   "<|hi|>": 50276,
   "<|hr|>": 50291,
   "<|ht|>": 50339,
@@ -38,7 +38,6 @@
   "<|id|>": 50275,
   "<|is|>": 50311,
   "<|it|>": 50274,
-  "<|iw|>": 50279,
   "<|ja|>": 50266,
   "<|jw|>": 50356,
   "<|ka|>": 50329,

config.json

@@ -2,6 +2,7 @@
   "_name_or_path": "openai/whisper-small",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
+  "apply_spec_augment": false,
   "architectures": [
     "WhisperForConditionalGeneration"
   ],
@@ -11,6 +12,7 @@
     50257
   ],
   "bos_token_id": 50257,
+  "classifier_proj_size": 256,
   "d_model": 768,
   "decoder_attention_heads": 12,
   "decoder_ffn_dim": 3072,
@@ -26,6 +28,12 @@
   "forced_decoder_ids": null,
   "init_std": 0.02,
   "is_encoder_decoder": true,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
   "max_length": 448,
   "max_source_positions": 1500,
   "max_target_positions": 448,
@@ -36,7 +44,8 @@
   "scale_embedding": false,
   "suppress_tokens": [],
   "torch_dtype": "float32",
-  "transformers_version": "4.25.0.dev0",
+  "transformers_version": "4.27.0.dev0",
   "use_cache": true,
+  "use_weighted_layer_sum": false,
   "vocab_size": 51865
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48dffa4ec7101a3398c9fbae98e54ea9803f9b0e34bbc2c692fd2e1276e34c7e
-size 967099139
+oid sha256:5082a514e030da7b81b3a7595361d134a5e06abc4f8c1465180fe5e3b4d5fa6c
+size 967102601

runs/Mar14_13-26-46_ea87badfbc29/1678800665.4242449/events.out.tfevents.1678800665.ea87badfbc29.766.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:485b280c11593ab90dbb8a22e397b1c218b45ce74ccf10eb387faafe7cb5c026
+size 6016

runs/Mar14_13-26-46_ea87badfbc29/events.out.tfevents.1678800665.ea87badfbc29.766.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06c5374f4b08f21353808b9832b3655f2687ff17cdb9d3dda5c2c97411aaef45
+size 11285

special_tokens_map.json

@@ -22,7 +22,7 @@
     "<|hi|>",
     "<|fi|>",
     "<|vi|>",
-    "<|iw|>",
+    "<|he|>",
     "<|uk|>",
     "<|el|>",
     "<|ms|>",
@@ -124,7 +124,7 @@
   },
   "pad_token": "<|endoftext|>",
   "unk_token": {
-    "content": "",
+    "content": "<|endoftext|>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,

tokenizer_config.json

@@ -19,7 +19,6 @@
   },
   "errors": "replace",
   "model_max_length": 1024,
-  "name_or_path": "openai/whisper-small",
   "pad_token": null,
   "processor_class": "WhisperProcessor",
   "return_attention_mask": false,
@@ -27,7 +26,7 @@
   "tokenizer_class": "WhisperTokenizer",
   "unk_token": {
     "__type": "AddedToken",
-    "content": "",
+    "content": "<|endoftext|>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9e0e23bb3182d4b4d7d4c244d6a74432d874544a0c5ad5e96d78e390d26a78a
-size 3503
+oid sha256:b59af55abac613a34f27e87fe4eb34ea1e9386af94e3a0ed05443ea1ced45ece
+size 3707

vocab.json

@@ -314,6 +314,7 @@
   ";;": 35746,
   "<": 27,
   "</": 3433,
+  "<|endoftext|>": 50257,
   "=": 28,
   "=\"": 13114,
   "=\"#": 34106,