Training in progress, step 250

config.json

@@ -9,7 +9,7 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.0,
+  "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
   "codevector_dim": 256,
@@ -49,12 +49,12 @@
   "eos_token_id": 2,
   "feat_extract_activation": "gelu",
   "feat_extract_norm": "group",
-  "feat_proj_dropout": 0.0,
+  "feat_proj_dropout": 0.1,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
   "freeze_feat_extract_train": true,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
+  "hidden_dropout": 0.1,
   "hidden_size": 768,
   "initializer_range": 0.02,
   "intermediate_size": 3072,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99de6adc1ed23f8e7a96973bc50511a1cf7b9e8a5bb11d8cde688f5f4fcb8ca9
+oid sha256:449029681b6c8b8857a188b311bcf800e2ab051f4ae0904d13e687f19513a02b
 size 377629528

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8109ba1ae334962d45909a7d00a358144dc7cce552d122fc2db5e5fc86d19325
+oid sha256:7763d94b5ce05259db6475142f52d75f65ae8148230c216507d7c98df1d41c77
 size 4283